The human location estimation in indoor enjoinment with good accuracy is required in many health care, surviving, and monitoring activities. The available methodologies are not providing good estimation accuracy or based on making big change in the environment infrastructure. This book, describes the methodologies used for human localisation in an indoor environment based on the strength of radio frequency signal. The book contribution is development of three techniques for representing the signal strength map and extracting location of a user from these maps. A strategy based on a particle filter, fuzzy logic, and neural networks for localisation are evaluated. A series of experiments conducted in a real-life environment and the viability of using radio frequency signal strength for localisation has been investigated and future research directions also highlight it. The methods will help shed some light on this new and exciting localisation system, and should be especially useful to professionals in Communications and intelligent systems fields, or anyone else who may be considering utilizing real time built environment human location applications.

© 2020, Springer Science+Business Media, LLC, part of Springer Nature. This paper consists of two parts. The first presents a review of the literature on the use of augmented reality (AR) in the diagnosis and treatment of autistic children with a particular focus on the efficacy of AR in assisting autistic children who have communicative, social, sentiment, and attention deficit disorders. The review also investigated interactions between AR systems and children, taking into consideration the target behaviors that are selected from the child during treatment. Such modes were fully explored by taking into account the needs of the individual child in terms of achieving an improvement in their condition. Most significantly, the empirical information that was obtained from the reviewed works was evaluated according to some specific targeted attitudes and how each AR solution was utilized during treatment to achieve the fostering of such attitudes in order to identify the requirements for building an effective AR system. In addition, the review revealed the essential design features that can enable AR systems to achieve a high level of effectiveness in autism therapy. The review also covered the instruction that AR systems were noticed to execute, and focused on the significant characteristics that allow AR systems to accomplish degrees of efficacy in autism treatment. The review ends by classifying the various AR systems based on different criteria. The second part of the paper focuses on our new AR system as a case study. It covers the design considerations and decisions as well as the key features and appearance of the system. The paper concludes by making some recommendations for the further development of an AR system for application in the domain of child autism.

© 2019 Elsevier B.V. Multiscale inherent fuzzy entropy is an objective measurement of electroencephalography (EEG) complexity, reflecting the habituation of brain systems. Entropy dynamics are generally believed to reflect the ability of the brain to adapt to a visual stimulus environment. In this study, we explored repetitive steady-state visual evoked potential (SSVEP)-based EEG complexity by assessing multiscale inherent fuzzy entropy with relative measurements. We used a wearable EEG device with Oz and Fpz electrodes to collect EEG signals from 40 participants under the following three conditions: a resting state (closed-eyes (CE) and open-eyes (OE) stimulation with five 15-Hz CE SSVEPs and stimulation with five 20-Hz OE SSVEPs. We noted monotonic enhancement of occipital EEG relative complexity with increasing stimulus times in CE and OE conditions. The occipital EEG relative complexity was significantly higher for the fifth SSVEP than for the first SSEVP (FDR-adjusted p < 0.05). Similarly, the prefrontal EEG relative complexity tended to be significantly higher in the OE condition compared to that in the CE condition (FDR-adjusted p < 0.05). The results also indicate that multiscale inherent fuzzy entropy is superior to other competing multiscale-based entropy methods. In conclusion, EEG relative complexity increases with stimulus times, a finding that reflects the strong habituation of brain systems. These results suggest that multiscale inherent fuzzy entropy is an EEG pattern with which brain complexity can be assessed using repetitive SSVEP stimuli.

Stroke represents a major health problem in our society. One of the effects of stroke is foot drop. Foot drop (FD) is a weakness that occurs in specific muscles in the ankle and foot such as the anterior tibialis, gastrocnemius, plantaris and soleus muscles. Foot flexion and extension are normally generated by lower motor neurons (LMN). The affected muscles impact the ankle and foot in both downward and upward motions. One possible solution for FD is to investigate the movement based on the bio signal (myoelectric signal) of the muscles. Bio signal control systems like electromyography (EMG) are used for rehabilitation devices that include foot drop. One of these systems is function electrical stimulation (FES). This paper proposes new methods and algorithms to develop the performance of myoelectric pattern recognition (M-PR), to improve automated rehabilitation devices, to test these methodologies in offline and real-time experimental datasets. Label classifying is a predictive data mining application with multiple applications in the world, including automatic labeling of resources such as videos, music, images and texts. We combine the label classification method with the self-advised support vector machine (SA-SVM) to create an adapted and altered label classification method, named the label self-advised support vector machine (LSA-SVM). For the experimental data, we collected data from foot drop patients using the sEMG device, in the Metro Rehabilitation Hospital in Sydney, Australia using Ethical Approval (UTS HREC NO. ETH15-0152). The experimental results for the EMG dataset and benchmark datasets exhibit its benefits. Furthermore, the experimental results on UCI datasets indicate that LSA-SVM achieves the best performance when working together with SA-SVM and SVM. This paper describes the state-of-the-art procedures for M-PR and studies all the conceivable structures.

The global number of people over the age of 60 years is expected to increase from 970 million to 2.1 billion in 2050 and 3.1 billion in 2100. About 80% of the aging population will be in the developing countries. Aging population may suffer from various physical, cognitive, and social problems, due to aging process such as impairment of physical related functions (decreased mobility and walking speed, falls, frailty, decreased
walking speed, difficulties in basic, and instrumental activities of daily living), cognitive related functions (memory-related issues), sensory functions (hearing loss, cataracts and refractive errors, presbyopia, decreased vestibular function), behavioural and psychological disorders, social isolation issues, and poor quality of life. Over the period of the last
few decades, emerging technologies such as internet of things (IoT), artificial intelligence (AI), sensors, cloud computing, wireless communication technologies, and assistive robotics have given the vision to develop various ambient or active assisted living (AAL) approaches for supporting an elderly people to live safely and independently in their living environment and participate in their daily and community activities, as well as supporting them to maintain their physical, mental health, and quality of their life. The aim of this paper is to review the use of Ambient or Active Assisted Living for older adults with physical, cognitive impairments, and their social participation.

OBJECTIVE:Clubfoot is a complex congenital three-dimensional foot deformity, which affects 150,000-200,000 newborn babies annually around the world. A good understanding of the alignment of the two osseous columns and the lower leg of the ankle and foot complex is essential for evaluating the severity of clubfoot. The purposes of this study were to (1) develop an automated three-dimensional (3D) surface model of severe clubfoot based on two-dimensional (2D) slices of computed tomography (CT) images, (2) evaluate the alignment of foot bones relative to the ankle in severe clubfoot, and (3) examine the structural changes in the shape of the clubfoot. PATIENTS AND METHODS:Two-dimensional CT image was taken from a four-year-old child with a severe clubfoot. Subsequently, an automated and detailed 3D surface model of the severe clubfoot was developed from the 2D images by using MATLAB software programming. Then, the x, y, and z coordinate angles were automatically calculated for each bone in the foot relative to the ankle (lower end of the tibia) to determine the orientations and relationships among the bones. RESULTS:The relative position or orientation of each bone of the foot to the ankle of the severe clubfoot was objectively measured which was used to determine the orientation of each bone in the foot. Among the x, y, and z axes of the interested tarsal bones, the z axis represents the smallest moment of inertia, and the results showed that the bones in the x axis shifted medially with higher relative angle. CONCLUSIONS:This 3D objective measurement method for assessing clubfoot can be used to determine and classify the severity of clubfoot, as well as evaluate and monitor the progress of the clubfoot intervention based on the relative position of the tarsal bones. The method can also be used to quantify the relationship between the tarsal bones of the foot and lower end of the tibia. In addition, angular measurements can be used to assess other pathological cond...

Copyright © 2019 Institute of Advanced Engineering and Science. All rights reserved. This paper introduces an automated system for skin cancer (melanoma) detection from Histo-pathological images sampled from microscopic slides of skin biopsy. The proposed system is a hybrid system based on Particle Swarm Optimization and Support Vector Machine (PSO-SVM). The features used are extracted from the grayscale image histogram, the co-occurrence matrix and the energy of the wavelet coefficients resulting from the wavelet packet decomposition. The PSO-SVM system selects the best feature set and the best values for the SVM parameters (C and γ) that optimize the performance of the SVM classifier. The system performance is tested on a real dataset obtained from the Southern Pathology Laboratory in Wollongong NSW, Australia. Evaluation results show a classification accuracy of 87.13%, a sensitivity of 94.1% and a specificity of 80.22%. The sensitivity and specificity results are comparable to those obtained by dermatologists.

© 2019 by the authors. Deep neural networks are successful learning tools for building nonlinear models. However, a robust deep learning-based classification model needs a large dataset. Indeed, these models are often unstable when they use small datasets. To solve this issue, which is particularly critical in light of the possible clinical applications of these predictive models, researchers have developed approaches such as virtual sample generation. Virtual sample generation significantly improves learning and classification performance when working with small samples. The main objective of this study is to evaluate the ability of the proposed virtual sample generation to overcome the small sample size problem, which is a feature of the automated detection of a neurodevelopmental disorder, namely autism spectrum disorder. Results show that our method enhances diagnostic accuracy from 84%-95% using virtual samples generated on the basis of five actual clinical samples. The present findings show the feasibility of using the proposed technique to improve classification performance even in cases of clinical samples of limited size. Accounting for concerns in relation to small sample sizes, our technique represents a meaningful step forward in terms of pattern recognition methodology, particularly when it is applied to diagnostic classifications of neurodevelopmental disorders. Besides, the proposed technique has been tested with other available benchmark datasets. The experimental outcomes showed that the accuracy of the classification that used virtual samples was superior to the one that used original training data without virtual samples.

© 2018 Institute of Advanced Engineering and Science. All rights reserved. Myoelectric pattern recognition (MPR) is used to detect user's intention to achieve a smooth interaction between human and machine. The performance of MPR is influenced by the features extracted and the classifier employed. A kernel extreme learning machine especially radial basis function extreme learning machine (RBF-ELM) has emerged as one of the potential classifiers for MPR. However, RBF-ELM should be optimized to work efficiently. This paper proposed an optimization of RBF-ELM parameters using hybridization of particle swarm optimization (PSO) and a wavelet function. These proposed systems are employed to classify finger movements on the amputees and able-bodied subjects using electromyography signals. The experimental results show that the accuracy of the optimized RBF-ELM is 95.71% and 94.27% in the healthy subjects and the amputees, respectively. Meanwhile, the optimization using PSO only attained the average accuracy of 95.53 %, and 92.55 %, on the healthy subjects and the amputees, respectively. The experimental results also show that SW-RBF-ELM achieved the accuracy that is better than other well-known classifiers such as support vector machine (SVM), linear discriminant analysis (LDA) and k-nearest neighbor (kNN).

Background: Congenital talipes equinovarus (CTEV) or clubfoot is a common pediatric congenital foot deformity that occurs 1 in 1,000 live births. Clubfoot is characterized by four types of foot deformities: hindfoot equinus; midfoot cavus; forefoot adductus; and hindfoot varus. A structured assessment method for clubfoot is essential for quantifying the initial severity of clubfoot deformity and recording the progress of clubfoot intervention. Aim: This study aims to develop a three-dimensional (3D) assessment method to evaluate the initial severity of the clubfoot and monitor the structural changes of the clubfoot after each casting intervention. In addition, this study explores the relationship between the thermophysiological changes in the clubfoot at each stage of the casting intervention and in the normal foot. Methods: In this study, a total of 10 clubfoot children who are <2 years old will be recruited. Also, the data of the unaffected feet of a total of 10 children with unilateral clubfoot will be obtained as a reference for normal feet. A Kinect 3D scanner will be used to collect the 3D images of the clubfoot and normal foot, and an Infrared thermography camera (IRT camera) will be used to collect the thermal images of the clubfoot. Three-dimensional scanning and IR imaging will be performed on the foot once a week before casting. In total, 6-8 scanning sessions will be performed for each child participant. The following parameters will be calculated as outcome measures to predict, monitor, and quantify the severity of the clubfoot: Angles cross section parameters, such as length, width, and the radial distance; distance between selected anatomical landmarks, and skin temperature of the clubfoot and normal foot. The skin temperature will be collected on selected areas (forefoot, mid foot, and hindfoot) to find out the relationship between the thermophysiological changes in the clubfoot at each stage of the casting treatment and in the normal foot. Ethics...

Feature extraction is taking a very vital and essential part of bio-signal processing. We need to choose one of two paths to identify and select features in any system. The most popular track is engineering handcrafted, which mainly depends on the user experience and the field of application. While the other path is feature learning, which depends on training the system on recognising and picking the best features that match the application. The main concept of feature learning is to create a model that is expected to be able to learn the best features without any human intervention instead of recourse the traditional methods for feature extraction or reduction and avoid dealing with feature extraction that depends on researcher experience. In this paper, Auto-Encoder will be utilised as a feature learning algorithm to practice the recommended model to excerpt the useful features from the surface electromyography signal. Deep learning method will be suggested by using Auto-Encoder to learn features. Wavelet Packet, Spectrogram, and Wavelet will be employed to represent the surface electromyography signal in our recommended model. Then, the newly represented bio-signal will be fed to stacked autoencoder (2 stages) to learn features and finally, the behaviour of the proposed algorithm will be estimated by hiring different classifiers such as Extreme Learning Machine, Support Vector Machine, and SoftMax Layer. The Rectified Linear Unit (ReLU) will be created as an activation function for extreme learning machine classifier besides existing functions such as sigmoid and radial basis function. ReLU will show a better classification ability than sigmoid and Radial basis function (RBF) for wavelet, Wavelet scale 5 and wavelet packet signal representations implemented techniques. ReLU will illustrate better classification ability, as an activation function, than sigmoid and poorer than RBF for spectrogram signal representation. Both confidence interval and Analysis of Va...

The extraction of the accurate and efficient descriptors of muscular activity plays an important role in tackling the challenging problem of myoelectric control of powered prostheses. In this paper, we present a new feature extraction framework that aims to give an enhanced representation of muscular activities through increasing the amount of information that can be extracted from individual and combined electromyogram (EMG) channels. We propose to use time-domain descriptors (TDDs) in estimating the EMG signal power spectrum characteristics; a step that preserves the computational power required for the construction of spectral features. Subsequently, TDD is used in a process that involves: 1) representing the temporal evolution of the EMG signals by progressively tracking the correlation between the TDD extracted from each analysis time window and a nonlinearly mapped version of it across the same EMG channel and 2) representing the spatial coherence between the different EMG channels, which is achieved by calculating the correlation between the TDD extracted from the differences of all possible combinations of pairs of channels and their nonlinearly mapped versions. The proposed temporal-spatial descriptors (TSDs) are validated on multiple sparse and high-density (HD) EMG data sets collected from a number of intact-limbed and amputees performing a large number of hand and finger movements. Classification results showed significant reductions in the achieved error rates in comparison to other methods, with the improvement of at least 8% on average across all subjects. Additionally, the proposed TSDs achieved significantly well in problems with HD-EMG with average classification errors of <5% across all subjects using windows lengths of 50 ms only.

The success of myoelectric pattern recognition (M-PR) mostly relies on the features extracted and classifier employed. This paper proposes and evaluates a fast classifier, extreme learning machine (ELM), to classify individual and combined finger movements on amputees and non-amputees. ELM is a single hidden layer feed-forward network (SLFN) that avoids iterative learning by determining input weights randomly and output weights analytically. Therefore, it can accelerate the training time of SLFNs. In addition to the classifier evaluation, this paper evaluates various feature combinations to improve the performance of M-PR and investigate some feature projections to improve the class separability of the features. Different from other studies on the implementation of ELM in the myoelectric controller, this paper presents a complete and thorough investigation of various types of ELMs including the node-based and kernel-based ELM. Furthermore, this paper provides comparisons of ELMs and other well-known classifiers such as linear discriminant analysis (LDA), k-nearest neighbour (kNN), support vector machine (SVM) and least-square SVM (LS-SVM). The experimental results show the most accurate ELM classifier is radial basis function ELM (RBF-ELM). The comparison of RBF-ELM and other well-known classifiers shows that RBF-ELM is as accurate as SVM and LS-SVM but faster than the SVM family; it is superior to LDA and kNN. The experimental results also indicate that the accuracy gap of the M-PR on the amputees and non-amputees is not too much with the accuracy of 98.55% on amputees and 99.5% on the non-amputees using six electromyography (EMG) channels.

Congenital talipes equinovarus (CTEV), also known as clubfoot, is common congenital orthopedic foot deformity in children characterized by four components of foot deformities: hindfoot equinus, hindfoot varus, midfoot cavus, and forefoot adduction. Although a number of conservative and surgical methods have been proposed to correct the clubfoot deformity, the relapses of the clubfoot are not uncommon. Several previous literatures discussed about the technical details of Ponseti method, adherence of Ponseti protocol among walking age or older children. However there is a necessity to investigate the relapse pattern, compliance of bracing, number of casts used in treatment and the percentages of surgical referral under two years of age for clear understanding and better practice to achieve successful outcome without or reduce relapse. Therefore this study aims to review the current evidence of Ponseti method (manipulation, casting, percutaneous Achilles tenotomy, and bracing) in the management of clubfoot under two years of age.Articles were searched from 2000 to 2015, in the following databases to identify the effectiveness of Ponseti method treatment for clubfoot: Medline, Cumulative Index to Nursing and Allied Health Literature (CINHAL), PubMed, and Scopus. The database searches were limited to articles published in English, and articles were focused on the effectiveness of Ponseti method on children with less than 2 years of age.Of the outcome of 1095 articles from four electronic databases, twelve articles were included in the review. Pirani scoring system, Dimeglio scoring system, measuring the range of motion and rate of relapses were used as outcome measures.In conclusion, all reviewed, 12 articles reported that Ponseti method is a very effective method to correct the clubfoot deformities. However, we noticed that relapses occur in nine studies, which is due to the non-adherence of bracing regime and other factors such as low income and social economic stat...

© 2017 ASTES Publishers. All rights reserved. Feature extraction is playing a significant role in bio-signal processing. Feature identification and selection has two approaches. The standard method is engineering handcraft which is based on user experience and application area. While the other approach is feature learning that based on making the system identify and select the best features suit the application. The idea behind feature learning is to avoid dealing with any feature extraction or reduction algorithms and to train the suggested model on learning features from input bio-signal by itself. In this paper, Self-Organizing Map (SOM) will be implemented as a feature learning technique to learn the model extract the features from the input data. Deep learning approach will be proposed by deploying SOM to learn features. In the proposed model, the raw data will be read then represented by using different signal representation as Spectrogram, Wavelet and Wavelet Packet. The newly represented data will be fed to self-organizing map layer to generate features, and finally, the performance of the suggested scheme will be evaluated by applying different classifiers such as Support Vector Machine, Extreme Learning Machine, Evolutionally Extreme Learning Machine and Discriminate Analysis Classification. Analysis of Variance (ANOVA) and confidence interval for different classifiers will be calculated. As an improving step for the results, classifier fusion layer will be implemented to select the most accurate result for both training and testing set. Classifier fusion layer led to a promising training and testing accuracies.

Clubfoot, known as congenital talipes equinovarus, is one of the complex paediatric foot deformity with the incidence of 1 in every 1000 live births. It consists of four complex foot abnormalities such as forefoot adductus, midfoot cavus, and hindfoot varus and ankle equinus. There are a number of surgical techniques (soft tissue releases, arthrodesis) used to correct clubfoot. However currently the conservative management (manipulation, serial casting, and braces) of clubfoot is considered as the best choice and it is widely accepted among orthopaedists. Clubfoot treated with surgical techniques might suffer various complications such as soft tissues contractures, neurovascular complications, infections, and shortening of the limbs. Although conservative method is generally considered as an effective method, it is still challenging to cure clubfoot in advance stages. Also, the classification of the initial severity of clubfoot is essential to evaluate the outcome of the treatment. In this review, the aim is to review the different types of conservative method and the assessment of clubfoot severity.

Abstract
OBJECTIVE:

Genu valgum is a common knee deformity in growing children. It alters the alignment of the lower extremity, body posture, and gait pattern of the children. Understanding of kinematic and kinetic parameters of gait in genu valgum is essential for planning and implementing the intervention to correcting the valgus deformity. The aim of this paper is to investigate the kinetic and kinematic gait differences in children with genu valgum.
PATIENTS AND METHODS:

A 13-year old girl with left side unilateral genu valgum and a closely matched healthy counterpart were recruited to compare the kinetic and kinematic parameters of their gait performances, and they were captured by The VICON motion analysis system.
RESULTS:

The results showed that the child with genu valgum had lower left and right knee angles (39.6˚; 30.2˚) and higher ankle angles (35.6˚; 28.4˚) than the healthy subject (64.2˚, 60.2˚). In addition, the child with genu valgum had lower moments on the left side of the knee (42.1 mm.N) than unaffected right knee (73.9 mm.N). Also, the ground reaction force was (0.7 N) lower in the affected knee of the child with genu valgum than the normal subject.
CONCLUSIONS:

This study revealed that there were decreased knee and ankle moments and lower knee and ankle ground reaction forces in the affected genu valgum extremity when compared with the healthy counterpart. These changes might be responsible for the altering gait pattern of the child with genu valgum.

We tackle the challenging problem of myoelectric prosthesis control with an improved feature extraction algorithm. The proposed algorithm correlates a set of spectral moments and their nonlinearly mapped version across the temporal and spatial domains to form accurate descriptors of muscular activity. The main processing step involves the extraction of the Electromyogram (EMG) signal power spectrum characteristics directly from the time-domain for each analysis window, a step to preserve the computational power required for the construction of spectral features. The subsequent analyses involve computing 1) the correlation between the time-domain descriptors extracted from each analysis window and a nonlinearly mapped version of it across the same EMG channel; representing the temporal evolution of the EMG signals, and 2) the correlation between the descriptors extracted from differences of all possible combinations of channels and a nonlinearly mapped version of them, focusing on how the EMG signals from different channels correlates with each other. The proposed Temporal-Spatial Descriptors (TSDs) are validated on EMG data collected from six transradial amputees performing 11 classes of finger movements. Classification results showed significant reductions (at least 8%) in classification error rates compared to other methods.

Long-term continuous patient monitoring is required in many health systems for monitoring and analytical diagnosing purposes. Most of monitoring systems have shortcomings related to their functionality and/or patient comfortably. Non-contact monitoring systems have been developed to address some of these shortcomings. One of such systems is non-contact physiological vital signs assessments for chronic heart failure (CHF) patients. This paper presents novel real-time demodulation technique and estimations algorithms for the non-contact physiological vital signs assessments for CHF patients based on a patented novel non-contact bio-motion sensor. A database consists of twenty CHF patients with New York Heart Association (NYHA) Heart Failure Classification Class II & III, whose underwent full Polysomnography (PSG) analysis for the diagnosis of sleep apnea, disordered sleep, or both, were selected for the study. The propose algorithms analyze the non-contact bio-motion signals and estimate the patient's respiratory and heart rates. The outputs of the algorithms are compared with gold-standard PSG recordings. Across all twenty CHF patients' recordings, the respiratory rate estimation median accuracy achieved 91.52% with median error of ±1.31 breaths per minute. The heart rate estimation median accuracy achieved 91.29% with median error of ±6.16 beats per minute. A potential application would be home continuous sleep and circadian rhythm monitoring.

Because of the functionality of an index finger, the disability of its motion in the modern age can decrease the person’s quality of life. As a part of rehabilitation therapy, the recognition of the index finger motion for rehabilitation purposes should be done properly. This paper proposes a novel recognition system of the index finger motion suing a cutting-edge method and its improvements. The proposed system consists of combination of feature extraction method, a dimensionality reduction and well-known classifier, Support Vector Machine (SVM). An improvement of SVM, Self-advise SVM (SA-SVM), is tested to evaluate and compare its performance with the original one. The experimental result shows that SA-SVM improves the classification performance by on average 0.63 %.

One of the main objectives of a successful lower limb robotic rehabilitation device is to obtain a smooth human machine interaction in different phases of gait cycle at the interaction point. The input (interaction force, Joint angle) and output (impedance) relationship of the control system is nonlinear. This paper proposes a fuzzy rule based controller to be used to control the interaction force at the patient exoskeleton interaction point. In achieving the objective, impedance, driver torque and angular velocity have been modulated in a way such that there is a reduction of interaction force. Minimum interaction force at the interaction point and tracking the defined gait trajectory with minimum error are set as benchmark to evaluate the performance in many tasks. In this paper there is an evaluation of what degree of impedance is ideal for what type of interaction force and joint angle to maintain a trajectory tunnel. This paper describes the control architecture of one Degree of freedom lower limb exoskeleton that has been specifically designed in order to ensure a proper trajectory control for guiding patient's limb along an adaptive reference gait pattern. The proposed methodology satisfies all the desired criteria for the device to be an ideal robotic rehabilitation device.

This paper presents the development of rehabilitation with biofeedback (RehaBio)
system for upper-limb rehabilitation that can be used to restore the upper-limb lost functions of patients who suffer from traumatic brain injury (TBI), spinal cord injury (SCI) or cerebrovascular accident (CVA), which generally result in paralysis on one side of the body. The system aims to close the gap in the requirements of one-to-one attention between physiotherapist and patient, to replace boring traditional upper-limb rehabilitation exercises and to reduce high healthcare cost. RehaBio is made up of three major modules: database module, rehabilitation exercise module and
biofeedback simulation module. Database module provides the information of the patients and their rehabilitation progress while rehabilitation exercise module provides with effective and motivated exercises based on augmented reality approach. Biofeedback simulation module in RehaBio serves two purposes: from physiotherapist point of view, it provides the tracking of biofeedback information of patient’s muscle performance and activities. From the patient’s point of view, it serves as a visual reflection of current activated muscles that create as an additional
motivation during training process. The effectiveness of the RehaBio system was evaluated by performing the experiments and provided with promising results.

This paper presents progress on treating children with Autism Spectrum Disorder (ASD) using Augmented Reality based games. The aim of these games is to enhance social interaction and hand-eye coordination in children with ASD thus easing them into becoming more comfortable around unfamiliar people. Colour detection and tracking and motion tracking concepts in augmented reality have been used to develop games for young children with ASD. The idea is that these games will encourage concentration and imagination from children through repetitive movement and visual feedback.

For the surviving stroke patients that are affected physically and mentally, they are required rehabilitation after the stroke. Rehabilitation can be quite expensive on the patient and their families. The augmented reality rehabilitation gaming system aims to decrease the dependency on supervised therapy. This paper presents two augmented reality games. The games focus on rehabilitating stroke patients affected with upper limb disabilities. The games simulate current physical therapy techniques in an interactive augmented environment. The benefits of using a gaming platform are to provide the user with increased motivation, as well as a cost effective rehabilitation solution. The games can be used with or without a hand held roller-ball device, which can change the movement and focus of the exercise.

Skin cancer is increasing and effect many people in different part of the world. Malignant melanoma as the deadliest type of skin cancer can be treated successfully if it detected early. Automatic detection is one of the most challenging research areas that can be used for early detection of such vital cancer. Over the last few years, many automatic diagnosis systems been suggested by different researchers targeting increasing of the diagnosis accuracy. This paper presents a quick review on the design of whole system and focus in preprocessing step of the automatic system. Preprocessing as the basis of automation system plays a vital role for accurate detection. This paper implements three techniques of contrast enhancement in the framework of three methodologies to find out the most effective one for further processing. The quality of resulted images in each methodology has been found based on testing the skin cancer images database using three image quality measurements.

Noise removing from an image is an important task in different applications such as medical which the noise free images could leads to
less error detection. Filtering as a tool for noise removal is concerned in this paper. The purpose is to compare the performance of five
filters - Median Filter, Adaptive Median Filter, Mean Filter, Gaussian Filter and Adaptive Wiener filter- for de-noising from Gaussian
noise, Salt & Pepper noise, Poisson noise and Speckle noise.

Automatic diagnostics of skin cancer is one of the most challenging problems in medical image processing. It helps physicians to decide whether a skin melanoma is benign or malignant. So, determining the more efficient methods of detection to reduce the rate of errors is a vital issue among researchers. Preprocessing is the first stage of detection to improve the quality of images, removing the irrelevant noises and unwanted parts in the background of the skin images. The purpose of this paper is to gather the preprocessing approaches can be used in skin cancer images. This paper provides good starting for researchers in their automatic skin cancer detections.

This paper proposes an automatic skin cancer (melanoma) classification system. The input for the proposed system is a set of images for benign and malignant skin lesions. Different image processing procedures such as smoothing and equalization are applied on these images to enhance their properties. Two segmentation methods are then used to identify the skin lesions before extracting the useful feature information from these images. This information is then passed to the classifier for training and testing. The features used for classification are coefficients created by Wavelet decompositions or simple wrapper Curvelets. Curvelets are known to be more suitable for the images that contain oriented textures and cartoon edges. The recognition accuracy obtained by the two layers back-propagation neural network classifier tested in this experiment is 58.44 % for the Wavelet based coefficients and 86.57 % for the Curvelet based ones

Physical disabilities which caused by neuromotor impairment due to Traumatic Brain Injury (TBI), Spinal Cord Injury (SCI) and Cerebrovascular Accident (CVA) affect the persons quality of life. Therefore, physical rehabilitations are required to be performed for the restoration of lost functions as a core treatment for such disabilities. However, the physical rehabilitations are too labor-intensive due to the nature of one-to-one attention in healthcare sectors. Moreover, this kind of injuries and accident cost over $10 billion per annum in healthcare sectors. To overcome above mentioned problems, this paper presents the development of intelligent biofeedback neuromotor rehabilitation system with low cost and motivational approach to close the gap in shortage of therapists, high healthcare cost of TBI, SCI and CVA. Our system designed for user motivation to perform the exercise longer and be used with minimum therapist supervision at home. The rehabilitation exercise aims to increase the upper limb range of motion, and strengthen the associate muscles. Our system utilized sEMG signals as a biofeedback. The users sEMG signals will attain and detect the therapist defined sEMG threshold level to display active muscle in real time during performing exercises. While the system works to retrain the elastic brain via fast recovery method, it will close the gap for the required information, by therapists, about monitoring and tracks the users muscle performance. The effectiveness of the proposed system has been evaluated by performing usability test.

Sleep apnea (SA) is recognized as the most important and common type of sleep disorders with several short term and long term side effects on health and prediction of sleep apnea events before they happened can help to prevent these side effects. There are several studies on automated SA detection but not too much works have been done on prediction of apnea's individual episodes. This paper investigated the application of artificial neural networks (ANNs) to predict sleep apnea. Three types of neural networks were investigated: Elman, RBF and feed-forward back propagation on data from 5 patients. Based on the obtained results, generally on all of experiments the best performance is obtained by the feed-forward neural network with average of Area-Under-Curve (AUC) statistic equal to But this superiority was not hold in all individual experiments and each of neural networks were be able to obtain the best result in some cases. This result showed the necessary of more investigation on methods such as dynamic neural networks selections instead of using a fixed model.

The Support Vector Machine (SVM) is one of the most popular machine learning algorithms for classification and regression. SVM displays outstanding performance when utilized in many applications. However, different approaches have been proposed in order

Image-based computer aided diagnosis systems have significant potential for screening and early detection of malignant melanoma. We review the state of the art in these systems and examine current practices, problems, and prospects of image acquisition, pre-processing, segmentation, feature extraction and selection, and classification of dermoscopic images. This paper reports statistics and results from the most important implementations reported to date. We compared the performance of several classifiers specifically developed for skin lesion diagnosis and discussed the corresponding findings. Whenever available, indication of various conditions that affect the technique’s performance is reported. We suggest a framework for comparative assessment of skin cancer diagnostic models and review the results based on these models. The deficiencies in some of the existing studies are highlighted and suggestions for future research are provided.

Skin Cancer; Segmentation; Diagnosis; Fuzzy; Thresholding; Level Sets

Current prosthetic and rehabilitation devices, used for those who are limbless or born with congenital defects or required rehabilitation, are difficult to use. The users have problems to adapt to their new hosts or receiving any bio-feedback despite reh

To get a compliant active exoskeleton controller, the force interaction controllers are mostly used in form of either the impedance or admittance controllers. The impedance or admittance controllers can only work if they are followed by either the force or the position controller respectively. These combinations place the impedance or admittance controller as high-level controller while the force or position controller as low-level controller. From the application point of view, the exoskeleton controllers are equipped by task controllers that can be formed in several ways depend on the aims. This paper presents the review of the control systems in the existing active exoskeleton in the last decade. The exoskeleton control system can be categorized according to the model system, the physical parameters, the hierarchy and the usage. These considerations give different control schemes. The main consideration of exoskeleton control design is how to achieve the best control performances. However, stability and safety are other important issues that have to be considered

Many kinds of upper limb rehabilitation systems have been developing for physically weak and/or injured patients to assist their daily life activities and promote their quality of life. Among those systems, EMG controlled rehabilitation systems provide the most effective and fastest ways to restore the lost functions due to such weakness and injuries. This paper presents the prediction of shoulder angle based on acquisition of surface electromyogram (sEMG) signals. Backpropagation neural network (BPNN) controller is developed to predict the angle of shoulder flexion/extension and abduction/adduction movements. Virtual Reality (VR) human model is developed to simulate the predicted shoulder angle which results from BPNN controller. Four sEMG signals are collected from user arm muscles and processed to extract their feature with root mean square (RMS) method. Then, the signals features directed to the neural network as input and the network predicts the angle of the shoulder joint as an output. The angles from BPNN drive the shoulder joint of the VR human model in virtual environment. Experiments were carried out to evaluate the effectiveness of the developed system and it was found that the constructed BPNN model and VR model can well represent the relationship between sEMG and shoulder joint angles and rotation. These positive results elaborate a move to design and develop the EMG controlled upper limb rehabilitation robot system to rehabilitate the physically weak person and paralysed patients.

Sleep apnea (SA) is one of the common sleep disorders. It has several consequences that can affect daily life activities. The common diagnose procedure is carried out through an overnight sleep test. The test usually includes of several bio-signals recordings that are used to detect this syndrome. The conventional approach of detecting the sleep apnea uses a manual analysis of most bio-signals to achieve reasonable accuracy. The manual process of this test, is highly cost and time consuming. This paper presents a novel automatic system for detecting and classifying apnea events by using just a few of bio-signals that are related to breathe defect. This method uses only the air flow, thoracic and abdominal respiratory movement as inputs for the system. The proposed technique consists of four main parts which are; signal segmentation, feature generation, feature selection and data reduction based on genetic SVM, and classification. Statistical analyzes on the attained results show efficiency of this system and its superiority versus previous methods even with more bio-signals as input.

Sleep Apnea (SA) is one of the most common andimportant part of sleep disorders. Unfortunately, sleep apneamay be going undiagnosed for years, because of the personsunawareness. The common diagnose procedure usuallyrequired an overnight sleep test. During the test, a recording ofmany biosignals, which related to breath, are obtained bypolysomnography machine to detect this syndrome. Themanual process for detecting the sleep Apnea by analysis therecording data is highly cost and time consuming. So, severalworks tried to develop systems that achieve this automatically.This paper proposes a genetic fuzzy approach for detectingApnea/Hypopnea events by using Air flow, thoracic andabdominal respiratory movement signals and Oxygen desaturation as the inputs. Results show efficiently of this approach.

In Australia, a major cause of disability is the stroke and it is the second highest cause of death after coronary heart disease. Studies have predicted that form 2008 to 2017 more than 0.5 million people is likely to suffer from stroke in Australia. In addition, after stroke 88% of the patients suffer from disability and stays at home. In this paper, a post stroke therapeutic device has been designed for hand motor function rehabilitation that a stroke survivor can use for bilateral movement practice. Out of twenty-one degrees of freedom of hand fingers, the prototype of the hand exoskeleton allowed fifteen degrees of freedom. The device is designed to be portable so that the user can engage in other activities while using the device. A prototype of the device is fabricated to provide complete flexion and extension motion of individual fingers of the left hand (impaired hand) based on the movements of the right hand (healthy hand) fingers. In addition, testing of the device on a healthy subject was conducted to validate if the design met the requirements.

One of the fundamental motivations for feature selection is to overcome the curse of dimensionality problem. This paper presents a novel feature selection method utilizing a combination of differential evolution (DE) optimization method and a proposed repair mechanism based on feature distribution measures. The new method, abbreviated as DEFS, utilizes the DE float number optimizer in the combinatorial optimization problem of feature selection. In order to make the solutions generated by the float-optimizer suitable for feature selection, a roulette wheel structure is constructed and supplied with the probabilities of features distribution. These probabilities are constructed during iterations by identifying the features that contribute to the most promising solutions. The proposed DEFS is used to search for optimal subsets of features in datasets with varying dimensionality. It is then utilized to aid in the selection of Wavelet Packet Transform (WPT) best basis for classification problems, thus acting as a part of a feature extraction process. Practical results indicate the significance of the proposed method in comparison with other feature selection methods.

One of the approaches enabling people with amputated limbs to establish some sort of interface with the real world includes the utilization of the myoelectric signal (MES) from the remaining muscles of those limbs. The MES can be used as a control input to a multifunction prosthetic device. In this control scheme, known as the myoelectric control, a pattern recognition approach is usually utilized to discriminate between the MES signals that belong to different classes of the forearm movements. Since the MES is recorded using multiple channels, the feature vector size can become very large. In order to reduce the computational cost and enhance the generalization capability of the classifier, a dimensionality reduction method is needed to identify an informative yet moderate size feature set. This paper proposes a new fuzzy version of the well known Fisher's Linear Discriminant Analysis (LDA) feature projection technique. Furthermore, based on the fact that certain muscles might contribute more to the discrimination process, a novel feature weighting scheme is also presented by employing Particle Swarm Optimization (PSO) for estimating the weight of each feature. The new method, called PSOFLDA, is tested on real MES datasets and compared with other techniques to prove its superiority.

The great majority of processes in distributed sensor networks, such as data aggregation methods, routing protocols, distributed voting schemes, misbehaviour detection systems, and so on, can only be accomplished via collaborative efforts of sensor nodes. However, the success of these processes is undermined by an attack termed controlled link establishment attack whose goal is to gain significant portion or even full control of distributed sensor networks through the controlled link establishment. This attack can be exemplified by node replication attack and key-swapping collusion attack based on their final attack goal. Thus countermeasures against the two latter attacks can be utilised as remedies for the controlled link establishment attack. Despite the fact that a growing body of such countermeasures has been proposed over recent years, each of them exposes its own limitations such as high performance overheads, unsound assumptions, and security weaknesses. Therefore, we propose a series of evolutionary schemes comprising naïve, adaptive, and extended schemes in this paper to overcome those limitations. The first two schemes are designed to be light-weight in performance at the cost of slightly weaker security robustness while the extended scheme obtains much greater security by trading off small performance. Theoretical analyses, simulations, and extensive comparison with other schemes have been conducted to demonstrate the plausibility of our schemes with respect to security features and performance overheads.

Developing accurate and powerful electromyogram (EMG) driven prostheses controllers that can provide the amputees with effective control on their artificial limbs, has been the focus of a great deal of research in the past few years. One of the major challenges in such research is extracting an informative subset of features that can best discriminate between the different forearm movements. In this paper, a new dimensionality reduction method, referred to as orthogonal fuzzy neighborhood discriminant analysis (OFNDA), is proposed as a response to such a challenge. Unlike existing attempts in fuzzy linear discriminant analysis, the objective of the proposed OFNDA is to minimize the distance between samples that belong to the same class and maximize the distance between the centers of different classes, while taking into account the contribution of the samples to the different classes. The proposed OFNDA is validated on EMG datasets collected from seven subjects performing a range of 5 to 10 classes of forearm movements. Practical results indicate the significance of OFNDA in comparison to many other feature projection methods (including locality preserving and uncorrelated variants of discriminant analysis) with accuracies ranging from 97.66% to 87.84% for 5 to 10 classes of movements, respectively, using only two EMG electrodes.

One of the most important tasks in any pattern recognition system is to find an informative, yet small, subset of features with enhanced discriminatory power. In this paper, a new neuro-fuzzy discriminant analysis based feature projection technique is presented based on a two stages hybrid of Neural Networks, optimized with Differential Evolution (DE), and a proposed Fuzzy Linear Discriminant Analysis (FLDA) technique. Although dimensionality reduction via FLDA can present a set of well clustered features in the reduced space, but like any version of the existing DAs it assumes that the original data set is linearly separable, which is not the case with many real world problems. In order to overcome this problem, the first stage of the proposed technique maps the initially extracted features in a nonlinear manner into a new domain, with larger dimensionality, in which the features are linearly separable. FLDA acts then on these linearly separable features to further reduce the dimensionality. The proposed combination, referred to as NFDA, is validated on a prosthetic device control problem with Electroencephalogram (EEG) datasets collected from 5 subjects achieving a maximum testing accuracy of 85.7% for a three classes of EEG based imaginations of movements.

The controller of a multifunction prosthetic hand usually employs a pattern recognition scheme to discriminate between the myoelectric signals (MES) from different classes of the forearm movements. The MES is recorded using a multichannel approach that makes the feature vector size very large. Hence a dimensionality reduction technique is needed to identify an informative moderate size feature set. This paper proposes a novel feature projection technique based on a combination of fisher linear discriminant analysis (LDA), fuzzy logic (FL), and differential evolution (DE) optimization technique. The new technique, DEFLDA, assigns different membership degrees to the data points in order to reduce the effect of overlapping points in the discrimination process. Furthermore, an optimized weighting scheme is presented in which certain weights are assigned to the features according to their contribution in the discrimination process. The proposed DEFLDA is tested on different datasets and compared with other projection techniques to prove its functionality.

Accurate and computationally efficient myoelectric control strategies have been the focus of a great deal of research in recent years. Although many attempts exist in literature to develop such strategies, deficiencies still exist. One of the major challenges in myoelectric control is finding an optimal feature set that can best discriminate between classes. However, since the myoelectric signal is recorded using multi channels, the feature vector size can become very large. Hence a dimensionality reduction method is needed to identify an informative, yet small size feature set. This paper presents a new feature selection method based on modifying the Particle Swarm Optimization (PSO) algorithm with the inclusion of Mutual Information (MI) measure. The new method, called BPSOMI, is a mixture of filter and wrapper approaches of feature selection. In order to prove its efficiency, the proposed method is tested against other dimensionality reduction techniques proving powerful classification accuracy.

Mean blood pressure (MBP) has high correlation with heart rate (HR), but such e relationship between them is ambiguous and nonlinear. This paper investigates establishing an accurate mathematical model to estimate MBP that is considering the influence of the stroke volume changing. Twenty three cases of MIMIC database till are employed; 12 cases for training and 11 cases for verification. The mean and standard deviation for all cases are calculated and compared with real results. Our suggested mathematical model achieved an encouragement results.

Path planning and obstacle avoidance are the two major issues in any navigation system. Wavefront propagation algorithm, as a good path planner, can be used to determine an optimal path. Obstacle avoidance can be achieved using possibility theory. Combining these two functions enable a robot to autonomously navigate to its destination. This paper presents the approach and results in implementing an autonomous navigation system for an indoor mobile robot. The system developed is based on a laser sensor used to retrieve data to update a two dimensional world model of therobot environment. Waypoints in the path are incorporated into the obstacle avoidance. Features such as ageing of objects and smooth motion planning are implemented to enhance efficiency and also to cater for dynamic environments.

© 2020, Springer Nature Switzerland AG. Face detection in thermal imaging has been used widely in human for different purposes such as surveillance, obtaining physiological reading: respiratory and heart rate from face region via thermal imaging. Physiological reading via infrared thermal imaging used in emotion and stress detection as well as polygraph analysis. In animal as general and cattle in specific, face region localized manually in order to obtain temperature for eyes, nose and mouth, which used for stress, diseases and inflammation detection. In order to develop a future automated system for monitoring health conditions in cattle, it required to detect the face region automatically. Based on author knowledge, there is no research done regarding face detection in cattle using infrared thermal images. Unlike the human, cattle keep roaming, which lead to a change in the face and body orientation. The main objective of this paper is proposing a new method for Multi-view face detection in cattle with accuracy enhancement by using three classifiers and temperature thresholding. Classifiers are established by using Histogram Oriented Gradient (HOG) as features and Support vector machine (SVM) for classification. The results show that the proposed algorithm is performing well in term of Specificity, Recall and F-measure and detection rate compare to the currently used method in the literature.

In order to interface the amputeeâs with the real world, the myoelectric signal (MES) from human muscles is usually utilized within a pattern recognition scheme as an input to the controller of a prosthetic device. Since the MES is recorded using multi channels, the feature vector size can become very large. In order to reduce the computational cost and enhance the generalization capability of the classifier, a dimensionality reduction method is needed to identify an informative moderate size feature set. This paper proposes a new fuzzy version of the well known Fisherâs Linear Discriminant Analysis (LDA) feature projection technique. Furthermore, based on the fact that certain muscles might contribute more to the discrimination process, a novel feature weighting scheme is also presented by employing Particle Swarm Optimization (PSO) for the weights calculation. The new method, called PSOFLDA, is tested on real MES datasets and compared with other techniques to prove its superiority.

One of the fundamental motivations for feature selection is to overcome the curse of dimensionality. A novel feature selection algorithm is developed in this chapter based on a combination of Differential Evolution (DE) optimization technique and statistical feature distribution measures. The new algorithm, referred to as DEFS, utilizes the DE float number optimizer in a combinatorial optimization problem like feature selection. The proposed DEFS highly reduces the computational cost while at the same time proves to present a powerful performance. The DEFS is tested as a search procedure on different datasets with varying dimensionality. Practical results indicate the significance of the proposed DEFS in terms of solutions optimality and memory requirements.

Feature set dimensionality reduction via Discriminant Analysis (DA) is one of the most sought after approaches in many applications. In this paper, a novel nonlinear DA technique is presented based on a hybrid of Artificial Neural Networks (ANN) and the Uncorrelated Linear Discriminant Analysis (ULDA). Although dimensionality reduction via ULDA can present a set of statistically uncorrelated features, but similar to the existing DAs it assumes that the original data set is linearly separable, which is not the case with most real world problems. In order to overcome this problem, a one layer feed-forward ANN trained with a Differential Evolution (DE) optimization technique is combined with ULDA to implement a nonlinear feature projection technique. This combination acts as nonlinear discriminant analysis. The proposed approach is validated on a Brain Computer Interface (BCI) problem and compared with other techniques.

The mean arterial pressure (MAP) is a very important cardiovascular parameter for physicians to diagnose various cardiovascular diseases. Many algorithms were used to estimate MAP with different accuracy. These algorithms used different factors, such as blood level, pulses, and external applied pressure, photo-plethysmography (PPG) signal features, heart rate (HR), and other factors. In addition, some natural-based techniques were employed to minimize the difference between estimated and measured blood pressure, as well as to measure blood pressure continuously.

The mean arterial pressure (MAP) is a very important cardiovascular parameter for physicians to diagnose various cardiovascular diseases. Many algorithms were used to estimate MAP with different accuracy. These algorithms used different factors, such as blood level, pulses, and external applied pressure, photo-plethysmography (PPG) signal features, heart rate (HR), and other factors. In addition, some natural-based techniques were employed to minimize the difference between estimated and measured blood pressure, as well as to measure blood pressure continuously. This article presents an algorithm to estimate MAP, utilizing the HR, Stroke Volume (SV), and Total Peripheral Resistance (TPR), with considering SV changing influence; this consideration is investigated mathematically, and by the Particle Swarm Optimization (PSO) technique.

Localization of human in indoor environment is one of the impartment issues in many applications. Building such real time system became fea- sible recently because of the increasing availability of mobile comput- ing devices that can be utilised in human localisation systems. Various types of systems and technologies have been developed for human local- ization. This chapter investigates the use of the radio frequency signal strength information extracted from a pre-installed wireless network as a localiser sensor to locate a mobile (user) human in built environment. Intelligent techniques (fuzzy logic and neural networks) based on RF are used to estimate the mobile user location in indoor environment... The obtained results, of busy working indoor environment, are very encouraging.

rapid increase in popularity in the past few years. The electromyography (EMG) signal, also referred to as the Myoelectric signal (MES), recorded at the surface of the skin, is one of the biosignals generated by the human body, representing a collection of electrical signals from the muscle fibre, acting as a physical variable of interest since it first appeared in the 1940s (Scott, 1984).

Bio-signals patterns analysis problems have enjoyed a rapid increase in popularity in the past few years. The electromyography (EMG) signal, also referred to as the Myoelectric signal (MES), recorded at the surface of the skin, is one of the biosignals generated by the human body, representing a collection of electrical signals from the muscle fibre, acting as a physical variable of interest since it first appeared in the 1940s (Scott, 1984). It was considered to be the main focus of scientists, and was advanced as a natural approach for the control of prosthesis, since it is utilising the electrical action potential of the residual limb’s muscles remaining in the amputee’s stump (which still has normal innervations, and thus is subject to voluntary control) as a control signal to the prosthesis—in other words, it allows amputees to use the same mental process to control their prosthesis as they had used in controlling their physiological parts; however, the technology in that time was not adequate to make clinical application viable. With the development of semiconductor devices technology, and the associated decrease in device size and power requirements, the clinical applications saw promise, and research and development increased dramatically.

© 2020, Springer Nature Switzerland AG. Surface Electrogastrogram (sEGG) examination is a noninvasive method based on recording the electrical signals of stomach muscles that can be used for investigation of a stomach wave propagation. sEGG records by placing electrodes on the abdominal skin. The recording of gastric dysrhythmias is an important tool for the clinician when patients have symptoms that suggest gastric dysfunction such as unexplained nausea, bloating, postprandial fullness, and early satiety. On the other hand, these upper gastrointestinal (GI) symptoms are nonspecific, and diseases or disorders of other organ such as esophagus, gallbladder, small bowel, colon, and non-GI diseases also can be considered. The progress of the EGG based application is very slow as EGG still suffers from several limitations. The aim of this study is to review the Electrogastrogram detection, analyzing methods and its application to enter the clinical world with recommendations for future study.

© 2019, Institute of Advanced Engineering and Science. All rights reserved. The detection of a hand movement beforehand can be a beneficent tool to control a prosthetic hand for upper extremity rehabilitation. To be able to achieve smooth control, the intention detection is acquired from the human body, especially from brain signal or electroencephalogram (EEG) signal. However, many constraints hamper the development of this brain-computer interface (BCI), especially for finger movement detection. Most of the researchers have focused on the detection of the left and right-hand movement. This article presents the comparison of various pattern recognition method for recognizing five individual finger movements, i.e., the thumb, index, middle, ring, and pinky finger movements. The EEG pattern recognition utilized common spatial pattern (CSP) for feature extraction. As for the classifier, four classifiers, i.e., random forest (RF), support vector machine (SVM), k-nearest neighborhood (kNN), and linear discriminant analysis (LDA) were tested and compared to each other. The experimental results indicated that the EEG pattern recognition with RF achieved the best accuracy of about 54%. Other published publication reported that the classification of the individual finger movement is still challenging and need more efforts to achieve better performance.

The classification technique is defined as a task of supervised learning aims to identify the set for which a new input element is classified. SVM is a classifier, which is simple in structure, easy for training and often applied in classification problems. In the paper introduced, we improved SVM model that employs local search (LS) to enhance the accuracy of the classification. The proposed approach was experimented by eleven standard benchmark datasets from UCI repository. The results show enhancement in classification accuracy of proposed SVM model in comparison with that of the conventional SVM model.

Our proposed new technique combines local search optimizer and support vector machine classifier as a reasonable solution to handle classification problems. SVM was utilized and then optimized by investing the LS optimization technique. Results of experiments drove using 10 benchmark datasets showed that the proposed LS-SVM outperformed the original SVM on all datasets. Also, when comparing the accuracy of LS-SVM to the accuracy of other approaches detailed in the literature, LS-SVM’s classification accuracy was the highest on most of eleven datasets.

© 2020, Springer Nature Switzerland AG. Malignant Melanoma in one of the most deadly skin cancer. In latest cancer diagnosis, pathologists examine biopsies for analysing cell morphology and tissue distribution for making diagnostic assessments. However, this process is quite subjective and has considerable variability. Automated computational diagnostic tools based on quantitative measures can help the diagnosis done by pathologists. The first and foremost step in automated histopathological image analysis is to properly segment the tissue structures such as nest, irregular distribution and melanocytic cells which indicate some disorder or potential cancer. This paper presents a novel technique for automatic segmentation of histopathological skin images, without user intervention. It is based on an innovative approach for utilizing the concepts of clustering and level set evolution. Firstly, the image is pre-processed to enhance the differentiating structural details. Then a novel orientation sensitive Fuzzy C mean clustering is used to generate the initial coarse segmentation and to calculate the controlling parameters for level set evolution. Later, refined fast level set based algorithm is used to finalize the segmentation process. Experimental analysis on a database of 150 histopathological images display the accuracy of the proposed method for detecting the melanocytic areas of the image, with true detection rate of 87.66% and Dice similarity coefficient of 0.88, when segmentation results are compared with images marked by expert pathologists.

© 2019 IEEE. The ongoing development of various lightweight and portable EEG signal acquisition devices provides the opportunity to implement home-based epilepsy monitoring. However, it is essential to apply a highly effective method to handle the limited computational power of such devices. In this paper, we propose a cost-effective method to classify epileptic seizure using stratified sampling technique. Additionally, to reduce the required computational power, this paper proposes a novel correlation and threshold-based feature selection algorithm. For evaluating the performance of our proposed method, five different classification algorithms are applied to classify the epileptic seizure from the reduced feature set. In our experiment, the random forest classifier shows the highest accuracy compared to other classifiers.

© 2020, Springer Nature Switzerland AG. The importance of early diagnosis of autism that leads to early intervention such thing shall increase the results of treating it. The Autism Spectrum Disorder (ASD) affects the children activities and caused difficulties in interaction, impairments in communication, delayed speech, and weak eye contact. These activities used as the base for ASD diagnosis decision. Children move their upper limb before some of the other activities. Moving upper limb can be based for ASD diagnosis decision for autistic children. Such paper examines diagnosing the ASD that depends on motioning the children’s upper-limb aged between two and four years based on executing specific procedures and machine learning. The approach that such study utilized is both (LDA) Linear Discriminant Analysis in order to elicit the features and (SVM) Support Vector Machines for classifying thirty children such study selected fifteen autistic children out of fifteen non-autistic children by testing the collected data that are collected from doing an easy task. The results of such study have accomplished an optimal sortation accuracy of 100% and the average accuracy of 93.8%. Such outcomes provide more proof of simple brachium motioning that might be utilized in sorting poor performance of autistic children precisely.

Surface Electromyogram (EMG) pattern recognition has long been utilized for controlling multifunctional myoelectric prostheses. In such an application, a number of EMG channels are usually utilized to acquire more information about the underlying activity of the remaining muscles in the amputee stump. However, despite the multichannel nature of this application, the extracted features are usually acquired from each channel individually, without consideration for the interaction between the different muscles recruited to achieve a specific movement. In this paper, we proposed an approach of spatial filtering, denoted as Range Spatial Filtering (RSF), to increase the number of EMG channels available for feature extraction, by considering the range of all possible logical combinations of each n channels. The proposed RSF method is then combined with conventional time-domain (TD) feature extraction, as an extension of the conventional single channel TD features that are heavily considered in this field. We then show how the addition of a new feature, specifically the minimum absolute value of the range of each two windowed EMG signals, can significantly reduce the different patterns misclassification rate achieved by conventional TD features (with and without our RSF method). The performance of the proposed method is verified on EMG data collected from nine transradial amputees (seven traumatic and two congenital), with six grip and finger movements, for three different levels of forces (low, medium, and high). The classification results showed significant reduction in classification error rates compared to other methods (nearly 10% for some individual TD features and 5% for combined TD features, with Bonferroni corrected p-values <; 0.01).

In this paper, we investigate the possibility of a machine-learning algorithm using the Support Victor Machine Regression (SVMR) to predict the motor functional recovery of moderate post stroke patients during their rehabilitation program. To train the model, we used the recorded electromyography (EMG) signals from the upper limb muscles of the patients during their initial rehabilitation sessions. Then we tested the trained model to predict the later muscles performance of the patient during the same sessions. The results of this pilot study were promising; data were, to some extent, predictable. We believe such research direction could be essential to motivate the patient to complete the designed rehabilitation program and can assist the therapist to innovate proper rehabilitation menu for individual patients.

© 2018 IEEE. Obstructive Sleep Apnea (OSA) occurs when obstruction happens repeatedly in the airway during sleep due to relaxation of the tongue and airway-muscles. Usual indicators of OSA are snoring, poor night sleep due to choking or gasping for air and waking up unrefreshed. OSA diagnosis is costly both in the monetary and timely manner. That is why many patients remain undiagnosed and unaware of their condition. Previous research has shown the link between facial morphology and OSA. In this paper, we investigate the application of deep learning techniques to diagnose the disease through depth map of human facial scans. Depth map will provide more information about facial morphology as compared to the plain 2-D color image. Even with very less amount of sample data, we are able to get around 69validation accuracy using transfer learning. We are predicting patients with above moderate > 15 or below moderate ≤ 15 OSA.

Copyright © 2019 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved. Infrared Thermography Technology (IRT) is a non-invasive method that has been used to calculate and display temperature as an Infrared thermal image. Infrared thermal images are used frequently to measure temperature remotely, and this temperature can be used as a health indicator for detecting diseases and inflammation in human and animal. In cattle, the rising temperature of the eye and nose region used for identifying stress and Bovine respiratory disease (BRD). In such applications, measuring temperature for nose and eye region is conducted manually. In this paper, a new automatic method is proposed for extracting the hottest regions from the face region, which may include eyes, nose and mouth. The proposed method involves face detection, thresholding, and blob refinement. The preliminary results show that the proposed algorithm is working well for localization and temperature measurement.

One of the most important electrophysiological signal is the Electromyography (EMG) signal, which is widely used in medical and engineering studies. This signal contains a wealth of information about muscle functions. Therefore, the EMG signal is becoming increasingly important and has started to be used in many applications like finger movement rehabilitation. However, an advanced EMG signal analysis method is required for efficient usage of such applications. This signal analysis can include signal detection, decomposition, processing, and classification. There are many approaches in studying the EMG signals, however, one of the important factor of analyzing is to get the most efficient and effective features that can be extracted from the raw signal. This paper presents the best feature extraction set compared to previous studies. Where eighteen well-known features algorithm has been tested using the sequential forward searching (SFS) method to get excellent classification accuracy in a minimum processing time. Among these novel features only four combinations have been selected with perfect results, which are; Hjorth Time Domain parameters (HTD), Mean Absolute Value (MAV), Root Mean Square (RMS) and Wavelet Packet Transform (WPT). The superiority of this feature set has been proven experimentally, and the results show that the classification accuracy could reach up to 99% to recognize the individual and combined for ten classes of finger movements using only two EMG channels.

© 2017 IEEE. Seagrass is an important component of the marine ecosystem and plays a vital role in preserving the water quality. The traditional approaches for sea grass identification are either manual or semi-automated, resulting in costlier, time consuming and tedious solutions. There has been an increasing interest in the automatic identification of seagrasses and this article provides a survey of automatic classification techniques that are based on machine learning, fuzzy synthetic evaluation model and maximum likelihood classifier along with their performance. The article classifies the existing approaches on the basis of image types (i.e. aerial, satellite, and underwater digital), outlines the current challenges and provides future research directions.

Recent studies indicate the limited clinical acceptance of myoelectric prostheses, as upper extremity amputees need improved functionality and more intuitive, effective, and coordinated control of their artificial limbs. Rather than exclusively classifying the electromyogram (EMG) signals, it has been shown that inertial measurements (IMs) can form an excellent complementary signal to the EMG signals to improve the prosthetic control robustness. We present an investigation into the possibility of replacing, rather than complementing, the EMG signals with IMs. We hypothesize that the enhancements achieved by the combined use of the EMG and IM signals may not be significantly different from that achieved by the use of Magnetometer (MAG) or Accelerometer (ACC) signals only, when the temporal and spatial information aspects are considered. A large dataset comprising recordings with 20 ablebodied and two amputee participants, executing 40 movements, was collected. A systematic performance comparison across a number of feature extraction methods was carried out to test our hypothesis. Results suggest that, individually, each of the ACC and MMG signals can form an excellent and potentially independent source of control signal for upper-limb prostheses, with an average classification accuracy of $\approx 93$% across all subjects. This study suggests the feasibility of moving from surface EMG to IM signals as a main source for upper-limb prosthetic control in real-life applications.

© 2017 IEEE. Sleep is crucial in our daily life as it plays a key role in our physical and mental health. It is important to monitor the sleep body orientations and movements due to its relationships to particular diseases, e.g., obstructive sleep apnea, insomnia or periodic limb movement disorder. Analyzing sleep body orientations also helps in determining sleep quality and irregular sleeping patterns. However, the current non-invasive sleep body orientations monitoring technologies are not well suited for long-term continuous monitoring due to its restrictions in mobility and comfort. This paper proposes a system that applies a features extraction process, utilizing wavelet packet decomposition, to extract features that describe the non-contact Doppler radar signatures caused by the body orientations. A database consisting of 24 chronic heart failure patients is selected for the training, validation and test of the non-contact body orientations prediction. These patients are diagnosed with New York Heart Association heart failure classification Class II & III and underwent full polysomnography analysis for the diagnosis of sleep apnea, disordered sleep, or both. The patients' data are randomly concatenated and partitioned into the ratio of 50% for 'Training', 15% for 'Validation' and 35% for 'Test. Across the 'Test dataset with total sleep duration of 65 hours, the body orientations prediction accuracy achieved a correct classification rate of 99.2% for 5 classes of 'Prone', 'Upright', 'Supine', 'Right and 'Left body orientations. The misclassification rate is 0.8%. A potential application would be non-contact continuous monitoring of nocturnal body orientations in the home.

© Springer Nature Singapore Pte Ltd. 2017. A variable translation wavelet neural network (VT-WNN) is a type of wavelet neural network that is able to adapt to the changes in the input. Different learning algorithms have been proposed such as backpropagation and hybrid wavelet-particle swarm optimization. However, most of them are time costly. This paper proposed a new learning mechanism for VT-WNN using random weights. To validate the performance of randomized VT-WNN, several experiments using benchmark data form UCI machine learning datasets were conducted. The experimental results show that RVT-WNN can work on a broad range of applications from the small size up to the large size with comparable performance to other well-known classifiers.

© 2018, Institute of Advanced Engineering and Science. All rights reserved. One of the challenges of the hand rehabilitation device is to create a smooth interaction between the device and user. The smooth interaction can be achieved by considering myoelectric signal generated by human's muscle. Therefore, the so-called myoelectric control system (MCS) has been developed since the 1940s. Various MCS's has been proposed, developed, tested, and implemented in various hand rehabilitation devices for different purposes. This article presents a review of MCS in the existing hand rehabilitation devices. The MCS can be grouped into main groups, the non-pattern recognition and pattern recognition ones. In term of implementation, it can be classified as MCS for prosthetic and exoskeleton hand. Main challenges for MCS today is the robustness issue that hampers the implementation of MCS on the clinical application.

© 2016 IEEE. There are wide verities of human movement possible that involves a range from the gait of the physically handicapped, the lifting of a load by a factory worker to the performance of a superior athlete. Output of the movement can be described by a large number of kinematic variables. Modeling each case with a muscle model is difficult. Intended action data can also be extracted from surface Electromyography (EMG) signal which may include intended torque, angle and impedance parameters of the knee joint dynamics. In this paper, Adaptive Neuro-Fuzzy Inference System (ANFIS) has been used in trying to estimate angle. As EMG signal is a function of angle, velocity and muscle activation level (load lifted), an adaptive machine learning technique is most desirable. Many different EMG signal intensity is possible at the same extension angle for different velocity of lower limb movement about knee joint. The EMG signal has been extracted from two different muscles and their patterns are very unique from velocity to velocity for entire range of extension angle. So a learning method of a Neural structure whose connections are based on rules is required to be able to estimate the angle at various speed about the knee joint as the slope of EMG signal intensity for each case of velocity varies significantly. The EMG signal has been collected from volunteer who has completed the knee joint extension in 15 Sec, 10 Sec, 8 Sec, 5 Sec, 3 Sec, 1 Sec, 0.5 Sec and 0.35 Sec respectively. RMS feature has been used to smooth the raw EMG signal. ANFIS is able to estimate angle adaptively although EMG pattern is changing with respect to speed. The simulation has shown experiment of comparative performance of angle estimation by different membership function and features.

© 2017 National Committee on Applied Mechanics. All Rights Reserved. Tremor is a neurological disorder characterized by involuntary oscillations. Difficulties associated with tremor in patients with Parkinson’s disease have motivated the researchers to work on developing various methods for tremor suppression. Active Force Control (AFC) method for tremor attenuation in human body parts is considered in this work. This paper proposes a new AFC system based on a piezoelectric actuator. A three-degree-of-freedom musculoskeletal model including wrist flexion-extension (FE), radial-ulnar deviation (RUD), and pronation supination (PS) is developed for studying tremor in the wrist joint. The musculoskeletal model for this study contains four muscles; extensor carpi radialis longus, extensor carpi ulnaris, flexor carpi ulnaris and flexor carpi radialis. Also, the muscle model is developed from the classic Hill-type muscle model. First, simulation of the tremor generation in the model is performed and then the performance of AFC system for suppressing wrist joint tremor is investigated. A single piezoelectric actuator is embedded in AFC system for controlling the behavior of the classic proportional-derivative controller. MATLAB Simulink is used to analyze the model. Results show that the AFC-based system with a piezoelectric actuator and a PD controller is very effective in suppressing the human hand tremor.

© 2017 IEEE. Infrared thermography technology (IRT) is a non-invasive method that has been used to measure the temperature from Infrared images. The measured temperature can be correlated to evaluate health status. Due to the ability of IRT for measuring temperature remotely, IRT has succeeded widely in detecting a different kind of diseases and inflammations in the animal. Many studies pointed out that eye is the best region for measuring body temperature. Recent work deals with automation of the measurement as a part of the computer-aided diagnosis system. However, such system still based on localization of the eye manually. This paper proposes a novel automated method for eyes localization in cattle. The proposed algorithm is work based on masking the face by modified ellipse detection algorithm. It also extends to remove the consideration for the position and orientation of target animal. Experimental results show that proposed algorithm is working well for localization purpose and temperature measurement as the final result.

Most of humankind feel sadness, tragic, feeling down from time to time; a few people encounter these emotions strongly, for long period of time and usually with no evident reason. Depression is not a low mood only; it's a genuine condition that affects the physical and mental health of the human. There are many studies that demonstrate a close association between depression and type 2 diabetes. Therefore, this paper aims to consolidate prediction of depression operation through the developing and applying the machine learning techniques. The supervised machine learning aims to construct a compact model of the allocation of class labels based on set of features to mimic the reality. The classification technique is used to give class labels to the subjects under testing based on values of the known prediction features, but the class label is unknown. In this paper state of art supervised learning classifiers have been used with modification to the used data. The results are very encouraging to use machine learning in the Prediction of Depression among Type 2 Diabetic Patients.

© 2017 IEEE. Development of automated diagnosis systems using machine learning and expert knowledge based data analysis requires effective automated learning models. However, models based on limited expert labeled training data can wrongly affect the results of diagnosis due to insufficient training knowledge acquired. On the other hand, getting more relevant analytical details from all the data used for training is an aspect that can enhance the efficiency of learning algorithms. This paper proposes a semi-advised training and classification algorithm that has the capability to effectively use limited labeled data along with abundant unlabeled data. It demonstrates the capability to use unlabeled data for training the algorithm by obtaining sufficient amount of information through incorporating an advised and/or partially supervised methodology. For comparative analysis, dermatological and histopathalogical images of skin cancer are used as experimental datasets. The proposed algorithm provided very impressive diagnosis outputs for both type of datasets in comparison to several other famous algorithms that are usually used in literature for classification.

© 2017 IEEE. Early diagnosis of children with autism means early intervention which is very important to increase child treatment outcomes. One of these signs is based on examination of child upper limb movements. This paper presents an evaluation of tasks methods for investigation of upper limb motor in children at High Risk (HR) for autism. This study examined three tasks and finds the upper limb motor type that discriminates the high risk of autism. These three tasks are: the first task is a throw a small ball into a transparent plastic followed by insert the ball into a clear tube open at both sides. In the second task, place a block into a large open box then place four similar blocks on a target block to make a tower and in the third task is put in a shape into a small fund with removable lids. The paper introduces using machine learning to discriminate between the children with high risk for autism. The feature extracting techniques Linear Discriminant Analysis (LDA) is used. After generating feature vectors, Support Vector Machine (SVM) and Extreme Learning Machine (ELM) are used for classification step. The results are very encouraging the maximum classification accuracy for a task that inserts the ball into a clear tube open at both sides with mean accuracy 75.0% and 81.67 with SVMs and ELM respectively.

© 2017 IEEE. Many physical rehabilitation treatments aim to improve the ability to use upper limbs for the stroke patients. The robots have the power to help in treatment or mitigation as noted by the outcomes of recent research studies. However, the availability of these devices in clinical and rehabilitation settings is still limited. The paper discusses the technical design aspects and implementation of the wearable robotics assistive device. Firstly, it discusses design requirements and challenges of some of the existing rehabilitation and assistive devices. The second part of the paper focuses on our proposed 3D printing robot hand assistive devices as a case study included decisions considerations, design decisions, and aspects of the designs. In addition, outline the future recommendations for such the devices development.

We tackle the challenging problem of myoelectric prosthesis control with an improved feature extraction algorithm. The proposed algorithm correlates a set of spectral moments and their nonlinearly mapped version across the temporal and spatial domains to form accurate descriptors of muscular activity. The main processing step involves the extraction of the Electromyogram (EMG) signal power spectrum characteristics directly from the time-domain for each analysis window, a step to preserve the computational power required for the construction of spectral features. The subsequent analyses involve computing 1) the correlation between the time-domain descriptors extracted from each analysis window and a nonlinearly mapped version of it across the same EMG channel; representing the temporal evolution of the EMG signals, and 2) the correlation between the descriptors extracted from differences of all possible combinations of channels and a nonlinearly mapped version of them, focusing on how the EMG signals from different channels correlates with each other. The proposed Temporal-Spatial Descriptors (TSDs) are validated on EMG data collected from six transradial amputees performing 11 classes of finger movements. Classification results showed significant reductions (at least 8%) in classification error rates compared to other methods.

Drop Foot (FD) is the inability to raising the foot due to the weakness of paralysis dorsiflexion muscle, this disability is caused by stroke. Recognize FD of the patient and provide a treatment as they needed is an important requirement for rehabilitation. One of the recognize techniques for FD event is based on using the surface electromyography (sEMG) signal. Utilizing sEMG signal can help to provide patient the specific rehabilitation treatment in specific time since it can detect the FD event before it happen. This paper has investigated the ability of Extreme Learning Machine (ELM) method to classify the sickness and healthy muscles on the leg, based on sEMG, that yield the FD. The performance is compared with Support Vector Machine (SVM) and Neural Network (NN). Classification accuracy with ELM is much better than SVM and NN giving the results with up to 97% classification accuracy using two channels on each side of the leg.

© 2016 IEEE. Crouch Gait (CG) can be observed in the hemiplegia persons after ischemic stroke. Walking with Crouch Gait (CG) shown a large gaits disorder. This paper explores the use of adaptive wavelet extreme learning machine (AW-ELM) to classifying different gait conditions for hemiplegia and healthy subjects. Three participants having a Crouch Gait problem with categories of Mild, Moderate, and Severe gait conditions, also, one Healthy person are used their data in this work. The recognition system extracting number of time and frequency domain features for dimensionality reduction. While for the classification stage, the common Extreme Learning Machine (ELM) classifiers are used. AW-ELM achieved maximum testing accuracy up to 91.149 % and with using majority vote post-processing the accuracy achieves 91.547 %.

© 2016 IEEE.The performance of the myoelectric pattern recognition system sharply decreases when working in various limb positions. The issue can be solved by cumbersome training procedure that can anticipate all possible future situations. However, this procedure will sacrifice the comfort of the user. In addition, many unpredictable scenarios may be met in the future. This paper proposed a new adaptive myoelectric pattern recognition using advance online sequential extreme learning (AOS-ELM) for classification of the hand movements to five different positions. AOS-ELM is an improvement of OS-ELM that can verify the adaptation validity using entropy. The proposed adaptive MPR was able to classify eight different classes from eleven subjects by accuracy of 95.42 % using data from one position. After learning the data from whole positions, the performance of the proposed system is 86.13 %. This performance was better than the MPR that employed original OS-ELM, but it was worse than the MPR that utilized the batch classifiers. Nevertheless, the adaptation mechanism of AOS-ELM is preferred in the real-time application.

© 2016 IEEE. Although Lower Limb Robotic Rehabilitation device exhibit a great prospect in the rehabilitation of impaired limb, yet it has not been widely applied to clinical rehabilitation of the patient with impairment. This is mostly due to insufficient bidirectional information interaction between exoskeleton and patient. The intended action data that can be extracted from surface electromyography (sEMG) signal may include the intended posture, intended torque, intended knee joint angle and intended desired impedance of the patient. Capturing intended knee joint torque from sEMG signal is one of the necessary parameter to achieve a smooth Human Machine Interaction force in a multilayer control mechanism. In this paper, a new technique to estimate Knee joint torque using SVM has been proposed that has used wavelet feature. The estimator is able to estimate required knee joint torque to lift 5kg, 12kg and 19kg weight using extensor and flexor muscles. Based on weight-Torque relationship, greater the weight, greater the torque is required to lift the weight. The estimator has classified required torque of 5kg, 12kg and 19kg with accuracy of 98.7296%, 86.0254% and 95.6443% respectively. The estimator can also be used to estimate torque about knee joint at different joint angle.

Although Lower Limb Robotic Rehabilitation device exhibit a great prospect in the rehabilitation of impaired limb, yet it has not been widely applied to clinical rehabilitation. This is mostly due to the insufficient bidirectional information interaction between exoskeleton and patient. In the shared control at the interaction point, it is very important that the deficiency of impaired lower limb in sharing the knee joint dynamics (Capturing of the intended action of the patient) is extracted beforehand to estimate as to how much assistance the robotic exoskeleton would provide. The intended action data that can be extracted from EMG signal may include the intended posture, intended torque, intended knee joint angle, intended knee joint torque and impedance parameter. In this paper, an application of Adaptive Network Based Fuzzy Inference System (ANFIS) has been proposed for proprioceptive feedback on the status of the interaction force at the patient robotic exoskeleton interaction point. ANFIS has been used to model the relationship between input and output. Interaction forces, rate of change in surface electromyography (EMG) signal are two inputs to ANFIS model and impedance parameters damping coefficients (also stiffness) is output. Impedance control law has damping as one of the tuning parameter. The resultant total torque is calculated from this law. The proposed model is able to estimate damping and demonstrate decent accuracy in modulating the knee joint dynamics to minimize the interaction force at the Patient Exoskeleton interaction point.

Although Lower Limb Robotic Rehabilitation device exhibit a great prospect in the rehabilitation of impaired limb, yet it has not been widely applied to clinical rehabilitation. This is mostly due to the insufficient bidirectional information interaction between exoskeleton and patient. In the shared control at the interaction point, it is very important that the deficiency of impaired lower limb in sharing the knee joint dynamics (Capturing of the intended action of the patient) is extracted beforehand to estimate as to how much assistance the robotic exoskeleton would provide. The intended action data that can be extracted from EMG signal may include the intended posture, intended torque, intended knee joint angle, intended knee joint torque and impedance parameter. In this paper, an application of Adaptive Network Based Fuzzy Inference System (ANFIS) has been proposed for proprioceptive feedback on the status of the interaction force at the patient robotic exoskeleton interaction point. ANFIS has been used to model the relationship between input and output. Interaction forces, rate of change in surface electromyography (EMG) signal are two inputs to ANFIS model and impedance parameters damping coefficients (also stiffness) is output. Impedance control law has damping as one of the tuning parameter. The resultant total torque is calculated from this law. The proposed model is able to estimate damping and demonstrate decent accuracy in modulating the knee joint dynamics to minimize the interaction force at the Patient Exoskeleton interaction point.

Although Lower Limb Robotic Rehabilitation device exhibit a great prospect in the rehabilitation of impaired limb, yet it has not been widely applied to clinical rehabilitation due to lack of identification of the system that can build a relationship relating knee joint dynamics to EMG signal. As a result, the insufficient bidirectional information interaction between exoskeleton and patient, the adaptive collaboration is very much absent. In the shared control situation at the interaction point, it is very important that the deficiency of impaired lower limb at knee joint dynamics (Capturing of the intended action of the patient) is extracted beforehand to estimate as to how much assistance need to be provided by the robotic exoskeleton. The intended action data that can be extracted from EMG signal may include the intended posture, intended torque, intended knee joint angle, intended knee joint torque and impedance parameter. In this paper, an application of Adaptive Network based Fuzzy Inference System (ANFIS) has been proposed to identify a proprioceptive feedback system which plays the role of inverse dynamics in the closed loop controlled Robotic Rehabilitation Device. The identified ANFIS inverse dynamic model updates the current status of the interaction force at the patient robotic exoskeleton interaction point. Interaction forces, rate of change in surface electromyography (EMG) signal, Extracted RMS (Root Mean Square) are three input patterns to ANFIS model and impedance parameters damping coefficients, stiffness are the output. The proposed model is able to estimate damping and demonstrate decent accuracy in modulating the knee joint dynamics to minimize the interaction force at the Patient Exoskeleton interaction point.

© 2016 IEEE. Tremor is a neurological disorder characterized by involuntary oscillations. Difficulties associated with tremor in patients with Parkinson's disease have motivated the researchers to work on developing various methods for tremor suppression. Active Force Control (AFC) method for tremor attenuation in human body parts is considered in this work. This paper proposes a new AFC system based on a piezoelectric actuator. A one-degree-of-freedom musculoskeletal model of the elbow joint with two links and one joint is developed. The model includes two muscles, biceps, and triceps as the flexor and the extensor of the elbow joint. First, simulation of the tremor generation in the model is performed and then the performance of AFC system for suppressing elbow joint tremor is investigated. A single piezoelectric actuator is embedded in AFC system for controlling the behaviour of the classic proportional-derivative controller. MATLAB Simulink is used to analyse the model. Results show that the AFC-based system with a piezoelectric actuator and a PD controller is very effective in suppressing the human hand tremor.

© 2016 IEEE. Feature extraction is playing a major role in bio signal processing. Feature identification and selection has two approaches. The common approach is engineering handcraft which is based on user experience and application area. While the other approach is feature learning that based on making the system identify and select the best features suit the application. The idea behind feature learning is to avoid dealing with any feature extraction or reduction algorithms and to train the suggested model on learning with avoiding the exposure to feature extraction which is mainly based on researcher experience. In this paper, Independent component analysis (ICA) will be implemented as a feature learning technique to learn the model extract the features from the input data. Deep learning approach will be proposed by implementing ICA to learn features. In the proposed model, the raw data will be read then represented by using different signal representation as Spectrogram, Wavelet and Wavelet Packet. Then, the new represented data will be fed to Independent component analysis layer to generate features and finally, the performance of the suggested scheme will be evaluated by applying different classifiers such as Support Vector Machine, Extreme Learning Machine and Discriminate Analysis. And As an improving step for the results, classifier fusion layer will be implemented to select the most accurate result for both training and testing set. Classifier fusion layer resulted in a promising training and testing accuracies. On the other side, Feature Selection is the process of selecting subset of features from a pool of features.

© 2016 IEEE. Feature extraction is playing a major role in bio signal processing. Feature identification and selection has two approaches. The common approach is engineering handcraft which is based on user experience and application area. While the other approach is feature learning that based on making the system identify and select the best features suit the application. The idea behind feature learning is to avoid dealing with any feature extraction or reduction algorithms and to train the suggested model. In this paper, principal component analysis (PCA) will be implemented as a feature learning technique to learn the model extract the features from the input data. Deep learning approach will be proposed by implementing PCA to learn features. In the proposed model, the raw data will be read then represented by using different signal representation as Spectrogram, Wavelet and Wavelet Packet. The new represented data will be fed to principal component analysis layer to generate features and finally, the performance of the suggested scheme will be evaluated by applying different classifiers such as Support Vector Machine, Extreme Learning Machine and Discriminate Analysis classification. As an improving step for the results, classifier fusion layer will be implemented to select the most accurate result for both training and testing set. Classifier fusion layer resulted in a promising training and testing accuracies. On the other side, Feature Selection is the procedure of picking out or choosing subgroup of features from a pool of features. Indices may be a good indicator or guider in determining which features to be combined with each other in order to avoid redundancy.

Computer aided classification of skin cancer images is an active area of research and different classification methods has been proposed so far. However, the supervised classification models based on insufficient labeled training data can badly influence the diagnosis process. To deal with the problem of limited labeled data availability this paper presents a semi advised learning model for automated recognition of skin cancer using histopathalogical images. Deep belief architecture is constructed using unlabeled data by making efficient use of limited labeled data for fine tuning done the classification model. In parallel an advised SVM algorithm is used to enhance classification results by counteracting the effect of misclassified data using advised weights. To increase generalization capability of the model, advised SVM and Deep belief network are trained in parallel. Then the results are aggregated using least square estimation weighting. The proposed model is tested on a collection of 300 histopathalogical images taken from biopsy samples. The classification performance is compared with some popular methods and the proposed model outperformed most of the popular techniques including KNN, ANN, SVM and semi supervised algorithms like Expectation maximization algorithm and transductive SVM based classification model.

Long-term continuous patient monitoring is required in many health systems for monitoring and analytical diagnosing purposes. It has been recognized that these types of monitoring systems have shortcomings related to patient comfort and/or functionality. Non-contact monitoring systems have been developed to address some of these shortcomings. One of such systems is non-contact physiological vital signs assessments for Chronic Heart Failure (CHF) patients. This paper presents a novel pulmonary ventilation model that defines the relationship between the intrapulmonary pressure and the chest displacement. A novel intrapulmonary pressure and tidal volume estimation algorithm is also proposed. A database consisting of twenty CHF patients with New York Heart Association (NYHA) Heart Failure Classification Class II & III; whose underwent full Polysomnography (PSG) analysis for diagnosis of sleep apnea, disordered sleep, or both, was selected for the verification of the proposed model and algorithm. The proposed algorithm analyzes the non-contact sensor data and estimate the patient’s intrapulmonary pressure and tidal volume. The output of the algorithm is compared with the gold-standard PSG recordings. Across all twenty CHF patients’ recordings with mean recorded sleep duration of 7.76 hours, the tidal volume estimation median accuracy achieved 83.13% with a median error of 57.32 milliliters. A potential application would be non-contact continuous monitoring of intrapulmonary pressure and tidal volume during sleep in the home.

© 2016 IEEE. Detection of autism in the beginning of infant life is very important to increase child treatment outcomes. There are many signs that help to diagnosis autism based on observation of great difficulty in communicating, delayed speech, and difficulty in forming relationships with other people. Attempts to discover other early symptoms and signs do not stop and tried to reach the early detection of autism. One of these signs is based on examination of child motor movements, such as kinematic for both gait and upper limb. This study aims to know how a basic upper-limb motion would be helpful to accurately classify High-Risk (HR) Infants for autism and infants with Low-Risk (LR) for autism. Also, the paper examined the task that throws a small ball into a transparent plastic tub followed by inserting the ball into a clear tube open at both ends performed by HR and LR subjects between the ages of 12 and 36 months. Then the data was analyzed by using Support Vector Machine (SVM) and Extreme Learning Machine (ELM). The results showed encouraging results in comparison to other state or art.

A robust myoelectric pattern-recognition-system
requires a system that should work in the real application as
good as in the laboratory. However, this demand should be
handled properly and rigorously to achieve a robust
myoelectric system. Electrode shift is an issue that usually
emerges when dealing with robustness issue. In daily life, the
placement of electrodes becomes a significant issue that can
downgrade the performance of the system. This paper
proposed a new way to overcome the robustness issue by
conducting an update to the system to anticipate changes in the
future such as electrode shift, improvement in muscle strength
or any other issue. Such update will be used to generate an
adaptation. The adaptation is done according to the user's need
by employing an online sequential extreme learning (OS-ELM)
to learn the training data chunk by chunk. OS-ELM enables
the myoelectric system to learn from a small number of data to
avoid cumbersome training process. The day-to-day
experiment shows that the proposed system can maintain its
performance on average accuracy around 85% whereas the
non-adaptive system could not.

Projecting a high dimensional feature into a low-dimensional feature without compromising the feature characteristic is a challenging task. This paper proposes a novel dimensionality reduction constituted from the integration of extreme learning machine (ELM) and spectral regression (SR). The ELM in the proposed method is built on the structure of the unsupervised ELM. The hidden layer weights are determined randomly while the output weight is calculated using the spectral regression. The flexibility of the SR that can take labels into consideration leads a new supervised dimensionality reduction called SRELM. Generally speaking, SRELM is an unsupervised system in term of ELM yet it is a supervised system in term of dimensionality reduction. In this paper, SRELM is implemented in the finger movement classification based on electromyography signals from two channels. The experimental results show that the SRELM can enhance the performance of its predecessor, spectral regression linear discriminant analysis (SRDA) because it has better class separability than SRDA. In addition, its performance is better than principal component analysis (PCA) and comparable to uncorrelated linear discriminant analysis (ULDA).

To capture the intended action of the patient and provide assistance as needed, the robotic rehabilitation device controller needs the intended posture, intended joint angle, intended torque and intended desired impedance of the patient. These parameters can be extracted from sEMG signal that are associated with knee joint. Thus an exoskeleton device requires a multilayer control mechanism to achieve a smooth Human Machine Interaction force. This paper proposes a method to estimate the required knee joint angles and associate parameters. The paper has investigated the feasibility of Extreme Learning Machine (ELM) as a estimator of the operation range of extension (0∘−90∘) and The performance is compared with Generalized Regression Neural Network (GRNN) and Neural Network (NN). ELM has performed relatively better than GRNN and NN

Capturing of the intended action of the patient and provide assistance as needed is required in the robotic rehabilitation device. The intended action data that can be extracted from surface Electromyography (sEMG) signal may include the intended posture, intended torque, intended knee joint angle and intended desired impedance of the patient. Utilizing such data to drive robotic assistive device like exoskeleton requires a multilayer control mechanism to achieve a smooth Human Machine Interaction force. It is very important that the controller for gait assistive device is able to extract as many information as possible from the patient muscle with impaired limb and predict different parameters associated with gait cycle. Joint kinematics and dynamics are important to be estimated as the Gait cycle of lower limb consists of flexion and extension postures at knee, hip and ankle joints respectively. This paper proposes a new classification and estimation technique of the lower limb joint kinematics and dynamics based on sEMG signal to predict specifically knee joint flexion and extension postures as well as Knee Joint angles of two postures. In the technique proposed, the feature data of raw sEMG data have been filtered with a second order digital filter and then input to train the Neural Network (NN) and to Generalized Regression Neural Network (GRNN) model to estimate the angle of flexion and extension. The GRNN and NN have been tested with RMS, LOG, MAV, IAV, Hjorth, VAR and MSWT features. GRNN with Multi scale Wavelet Transform (MSWT) feature has ensured 1.5704 Mean Square Error which is very promising accuracy. The SVM has been used to predict postures (flexion and extension). The SVM also has classified flexion and extension with accuracy over 95%.

Continuous prediction of dynamic joint angle from surface electromyography (sEMG) signal is one of the most important applications in rehabilitation area for stroke
survivors as these can directly reflect the user motor intention. In this study, new shoulder joint angle prediction method in real-time based on the biosignal: sEMG is proposed. Firstly, sEMG to muscle activation model is built up to extract the user intention from contracted muscles and then feed into the extreme learning machine (ELM) to estimate the angle in realtime continuously. The estimated joint angle is then compare with the webcam captured joint angle to analyze the effectiveness of the proposed method. The result reveals that correlation coefficient between actual angle and estimated angle is as high as 0.96 in offline and 0.93 in online mode. In addition, the processing time for the estimation is less than 32ms in both cases which is within the semblance of human natural movements. Therefore, the proposed method is able to predict the user intended movement very well and naturally and hence, it is suitable for real-time applications.

The application of neuroscience methods to analyze and understand preference formation and decision making in marketing tasks has recently gained research attention. The key contribution of this paper is to complement the advancement of traditional consumer research through the investigation of the event-related potentials (ERPs) associated with preferences elicited during a discrete choice experiment (DCE). Five subjects participated in the experiment as they chose their preferred computer background image from a set of images with different colors and patterns. Emotiv EPOC, a commercial wireless Electroencephalogram (EEG) headset with 14 channels, was utilized to collect EEG signals from the subjects while making one hundred and fifty choice observations. The collected EEG signals were filtered and cleaned from artifacts before being epoched into segments of 1000 msec each for ERP analysis. When observing the average of EEG epochs, collected while the subjects chose their preferred background images, there was a clear P300-ERP component with its largest power shown at the left frontal channel (F3 from the international 10-20 system). A significant difference was revealed between the average ERP potential on F3 during the epochs that coincided with the images containing the preferred objects against that coinciding with the images that did not contain the objects of interest (with p <0.01). A clear N400-ERP component on the parietal lobe sensor at P7 was also revealed to be significantly related to the difference in absolute preference (with p <0.02). Our experimental results also showed that there was a negative relationship between the speed of the decision and the difference in preference for the objects in the decision.

This paper proposes a pattern recognition model for classification. Adaptive differential evolution based feature selection is used for dimensionality reduction and a new advised version of support vector machine is used for evaluation of selected features and for the classification. The tuning of the control parameters for differential evolution algorithm, parameter value optimization for support vector machine and selection of most relevant features form the datasets all are done together. This helps in dealing with their interdependent effect on the overall performance of the learning model. The proposed model is tested on some latest machine learning medical datasets and compared with some well-developed methods in literature. The proposed model provided quite convincing results on all the test datasets.

Automated detection of cancerous tissue in histopathological images is a big challenge. This work proposed a new pattern recognition method for histopathological image analysis for identification of cancerous tissues. It comprised of feature extraction using a combination of wavelet and intensity based statistical features and autoregressive parameters. Moreover, differential evolution based feature selection is used for dimensionality reduction and an efficient self-advised version of support vector machine is used for evaluation of selected features and for the classification of images. The proposed system is trained and tested using a dataset of 150 histopathological images and showed promising comparative results with an average diagnostic accuracy of 89.1%.

Early diagnosis of skin cancer is one of the greatest challenges due to lack of experience of general practitioners (GPs). This paper presents a clinical decision support system aimed to save time and resources in the diagnostic process. Segmentation, feature extraction, pattern recognition, and lesion classification are the important steps in the proposed decision support system. The system analyses the images to extract the affected area using a novel proposed segmentation method H-FCM-LS. The underlying features which indicate the difference between
melanoma and benign lesions are obtained through intensity, spatial/frequency and texture based methods. For classification purpose, self-advising SVM is adapted which showed improved classification rate as compared to standard SVM. The presented work also considers analyzed performance of linear and kernel based SVM on the specific skin lesion diagnostic problem and discussed corresponding findings. The best diagnostic rates obtained through the proposed method are around 90.5 %.

Automated diagnosis of skin cancer is an active
area of research with different classification methods proposed
so far. However, classification models based on insufficient
labeled training data can badly influence the diagnosis process
if there is no self-advising and semi supervising capability in
the model. This paper presents a semi supervised, self-advised
learning model for automated recognition of melanoma using
dermoscopic images. Deep belief architecture is constructed
using labeled data together with unlabeled data, and fine
tuning done by an exponential loss function in order to
maximize separation of labeled data. In parallel a self-advised
SVM algorithm is used to enhance classification results by
counteracting the effect of misclassified data. To increase
generalization capability and redundancy of the model,
polynomial and radial basis function based SA-SVMs and Deep
network are trained using training samples randomly chosen
via a bootstrap technique. Then the results are aggregated
using least square estimation weighting. The proposed model is
tested on a collection of 100 dermoscopic images. The variation
in classification error is analyzed with respect to the ratio of
labeled and unlabeled data used in the training phase. The
classification performance is compared with some popular
classification methods and the proposed model using the deep
neural processing outperforms most of the popular techniques
including KNN, ANN, SVM and semi supervised algorithms
like Expectation maximization and transductive SVM.

Long term continuous patient monitoring is required in many health systems for monitoring and analytical diagnosing purposes. Most of monitoring systems had shortcomings related to their functionality or patient comfortably. Non-contact continuous monitoring systems have been developed to address some of these shortcomings. One of such systems is non-contact physiological vital signs assessments for chronic heart failure (CHF) patients. This paper presents a novel automated estimation algorithm for the non-contact physiological vital signs assessments for CHF patients based on a patented novel non-contact biomotion sensor. A database consists of twenty CHF patients with New York Heart Association (NYHA) heart failure Classification Class II & III, whose underwent full Polysomnography (PSG) analysis for the diagnosis of sleep apnea, disordered sleep, or both, were selected for the study. The patients mean age is 68.89 years, with mean body weight of 86.87 kg, mean BMI of 28.83 (obesity) and mean recorded sleep duration of 7.78 hours. The propose algorithm analyze the non-contact biomotion signals and estimate the patients' respiratory and heart rates. The outputs of the algorithm are compared with gold-standard PSG recordings. Across all twenty patients' recordings, the respiratory rate estimation median accuracy achieved 92.4689% with median error of ±1.2398 breaths per minute. The heart rate estimation median accuracy achieved 88.0654% with median error of ±7.9338 beats per minute. Due to the good performance of the propose novel automated estimation algorithm, the patented novel non-contact biomotion sensor can be an excellent tool for long term continuous sleep monitoring for CHF patients in the home environment in an ultra-convenient fashion.

Severe melanoma is potentially life-threatening. A novel methodology for automatic feature extraction from histopathological images and subsequent classification is presented. The proposed automated system uses a number of features extracted from images of skin lesions through image processing techniques which consisted of a spatially winner and adaptive median filter then applied Gabor filter bank to improve
diagnostic accuracy. Histogram equalization to enhance the contrast of the images prior to segmentation is used. Then, a wavelet approach is used to extract the features; more specifically Wavelet Packet Transform (WPT).This article introduces a novel melanoma detection strategy using a hybrid particle swarm - based support vector machine (SVM–WLG – PSO) technique. The extracted features are reduced by using a
particle swarm optimization (PSO), this was used to optimize the SVM parameters as a feature selection and finally, the obtained statistics are fed to a support vector machine (SVM) binary classifier to diagnose skin biopsies from patients as either malignant melanoma or benign nevi. The obtained classification accuracies show better performance in comparison to similar approaches for feature extraction. The proposed system is able to achieve one of the best results with classification accuracy of 87.13%, sensitivity of 94.1% and specificity of 80.22%.

a novel methodology for automatic feature extraction from histo-pathological images and subsequent classification is presented. The proposed automated system use a number of features extracted from images of skin lesions through image processing techniques which consisted of a spatially winner and adaptive median filter then applied Gabor filter bank to improve diagnostic accuracy. Histogram equalization to enhance the contrast of the images prior to segmentation is used. The extracted features are reduced by using sequential feature selection and finally, the obtained statistics are fed to a support vector machine (SVM) binary classifier to diagnose skin biopsies from patients as either malignant melanoma or benign nevi. The obtained classification accuracies show better performance in comparison to similar approaches for feature extraction. The proposed system is able to achieve a good result with classification accuracy of (81)%, sensitivity of(76)% and specificity of (lOO)%and 17 times faster than some of the reported results.

This paper proposes a new structure of wavelet extreme learning machine, i.e. an adaptive wavelet extreme learning machine (AW-ELM) for finger motion recognition using only two EMG channels. The adaptation mechanism is performed by adjusting the wavelet shape based on the input information. The performance of the proposed method is compared to ELM using wavelet (W-ELM) and sigmoid (Sig-ELM) activation function. The experimental results demonstrate that the proposed AW-ELM performs better than W-ELM and Sig-ELM.

An Accurate finger movement recognition is required in many robotic prosthetics and assistive hand devices. The use of a small number of Electromyography (EMG) channels for classifying the finger movement is a challenging task. This paper proposes the a novel recognition system which employs Spectral Regression Discriminant Analysis (SRDA) for dimensionality reduction, kernel-based Extreme Learning Machine (ELM) for classification and the majority vote for the classification smoothness. Particle Swarm Optimization is used to optimize the kernel-based ELM. Three hybridization with three kernels, radial basis function (SRBF-ELM), linear (SLIN-ELM), and polynomial (SPOLY-ELM) are introduced. The experimental results show that SRBF-ELM significantly outperforms SLIN-ELM but not too much different compared to SPOLY-LIN. Moreover, PSO is able to optimize the three systems by giving the accuracy more than 90% with the highest accuracy is ~94%.

The use of a small number of surface electromyography (EMG) channels on the transradial amputee in a myoelectric controller is a big challenge. This paper
proposes a pattern recognition system using an extreme learning machine (ELM) optimized by particle swarm optimization (PSO). PSO is mutated by wavelet function to avoid trapped in a local minima. The proposed system is used to
classify eleven imagined finger motions on five amputees by using only two EMG channels. The optimal performance of wavelet-PSO was compared to a grid-search method and standard PSO. The experimental results show that the proposed system is the most accurate classifier among other tested
classifiers. It could classify 11 finger motions with the average accuracy of about 94 % across five amputees.

One of the main objectives of a successful lower
limb robotic rehabilitation device is to obtain a smooth human
machine interaction in different phase of gait cycle. The new
concept in robotics rehabilitation is a “cooperative patient
strategy” meaning patient’s voluntary efforts are taken into
account rather than imposing any predefined movements or
inflexible strategies. The term cooperative is defined to include
compliance of robot as it behaves soft and gentle. It only reacts
to muscular effort, interactive because there is a bidirectional
exchange of energy and information between robot and patient.
The control of trajectory is shared by robot and patient to
complete gait cycle. In this paper an effort has been made to
establish a control law to tune the inertia, damping and
stiffness which in turn produce the desired trajectory
impedance for smooth human machine interaction. The gain
margin and phase margin in bode plot of our system is positive
and hence a stable system.

this is evident that training duration is a key factor for
a successful therapy. Robot supported therapy can improve the
rehabilitation allowing more intensive training. This paper
presents the kinematic, the control architecture and benchmark
criteria to evaluate the performance of Robotic Rehabilitation
Devices (RRD). Equipped with position, force and impedance
controller, the proposed RRD can deliver the patient cooperative
lower limb therapy taking into account the patient activity and
supporting him/her only as much as needed[1]. One of the main
objectives of a successful lower limb robotic rehabilitation device
is to obtain a smooth human machine interaction in different
phase of gait cycle at the interaction point (haptic behavior). The
input (interaction force, Joint angle, rate of change of interaction
force) and output (impedance, Δτ) relationship of the control
system is nonlinear. This paper proposes a fuzzy rule based
controller to be used to control the interaction force at the patient
exoskeleton interaction point. In achieving the objective,
impedance, driver torque and angular velocity have been
modulated in a way such that there is a reduction of interaction
force. Minimum interaction force at the interaction point and
tracking the defined gait trajectory with minimum error are set
as the benchmark to evaluate the performance in many tasks. In
this paper there is an evaluation of what degree of impedance is
ideal for what type of interaction force and joint angle to
maintain a trajectory tunnel. This paper describes the control
architecture of one Degree of freedom lower limb exoskeleton
that has been specifically designed in order to ensure a proper
trajectory control for guiding patient’s limb along an adaptive
reference gait pattern[2]. The proposed methodology satisfies all
the desired criteria to be an ideal robotic rehabilitation device.

This paper presents design and development of real time biosignal-driven illusion system: Augmented Reality based Illusion System (ARIS) for upper limb motor rehabilitation. ARIS is a hospital / home based self-motivated whole arm rehabilitation system that aims to improve and restore the lost upper limb functions due to Cerebrovascular Accident (CVA) or stroke. Taking the advantage of human brain plasticity nature, the system incorporates with number of technologies to provide fast recovery by re-establishing the neural pathways and synapses that able to control the mobility. These technologies include Augmented Reality (AR) where illusion environment is developed, computer vision technology to track multiple colors in real time, EMG acquisition system to detect the user intention in real time and 3D modelling library to develop Virtual Arm (VA) model where human biomechanics are applied to mimic the movement of real arm. The system operates according to the user intention via surface electromyography (sEMG) threshold level. In the case of real arm cannot reach to the desired position, VA will take over the job of real arm to complete the exercise. The effectiveness of the developed ARIS has evaluated via questionnaire, graphical and analytical measurements which provided with positive results.

This paper proposes a novel upper extremity rehabilitation system with virtual arm illusion. It aims for fast recovery from lost functions of the upper limb as a result of stroke to provide a novel rehabilitation system for paralyzed patients. The system is integrated with a number of technologies that include Augmented Reality (AR) technology to develop game like exercise, computer vision technology to create the illusion scene, 3D modeling and model simulation, and signal processing to detect user intention via EMG signal. The effectiveness of the developed system has evaluated via usability study and questionnaires which is represented by graphical and analytical methods. The evaluation provides with positive results and this indicates the developed system has potential as an effective rehabilitation system for upper limb impairment.

A great challenge of research and development activities have recently highlighted in segmenting of the skin cancer images. This paper presents a novel algorithm to improve the segmentation results of level set algorithm with skin cancer images. The major contribution of presented algorithm is to simplify skin cancer images for the computer aided object analysis without loss of significant information and to decrease the required computational cost. The presented algorithm uses k-means clustering technique and explores primitive segmentation to get initial label estimation for level set algorithm. The proposed segmentation method provides better segmentation results as compared to standard level set segmentation technique and modified fuzzy cmeans clustering technique.

Accurate segmentation of skin lesion is one of the
most important step for automated diagnosis of skin cancer.
Various characteristics of skin lesions and intensity variations
in images can make it a highly challenging task. A new
histogram analysis based fuzzy C mean thresholding method is
presented here. It unifies the advantages of soft and hard
thresholding algorithms along with reducing the computational
complexity. Appropriate threshold value can be calculated even
in the presence of abrupt intensity variations. This algorithm
shows significantly improved performance for the segmentation
of skin lesions. Experimental verification is done on a large set
of skin lesion images having almost all types of expected
artifacts that may badly affect the segmentation results.
Performance evaluation is done by comparing the diagnosis
results based on this method with other state of the art
thresholding methods. Results show that the proposed
approach performs reasonably well and can form a basis of
expert diagnostic systems for skin cancer.

Melanoma is the most deathly of all skin cancers but early diagnosis
can ensure a high degree of survival. Early diagnosis is one of the greatest challenges
due to lack of experience of general practitioners (GPs). In this paper we
present a clinical decision support system designed for general practitioners,
aimed at saving time and resources in the diagnostic process. Segmentation,
pattern recognition, and change detection are the important steps in our approach.
This paper also investigates the performance of Artificial Neural Network
(ANN) learning algorithms for skin cancer diagnosis. The capabilities of
three learning algorithms i.e. Levenberg-Marquardt (LM), Resilient Back propagation
(RP), Scaled Conjugate Gradient (SCG) algorithms in differentiating
melanoma and benign lesions are studied and their performances are compared.
The results show that Levenberg-Marquardt algorithm was quick and efficient
in figuring out benign lesions with specificity 95.1%, while SCG algorithm
gave better results in detecting melanoma at the cost of more number of epochs
with sensitivity 92.6%.

Early diagnosis of skin cancer is one of the greatest challenges due to
lack of experience of general practitioners (GPs). This paper presents a clinical
decision support system aimed to save lives, time and resources in the early diagnostic
process. Segmentation, feature extraction, and lesion classification are
the important steps in the proposed system. The system analyses the images to
extract the affected area using a novel proposed segmentation method H-FCMLS.
A set of 45 texture based features is used. These underlying features which
indicate the difference between melanoma and benign images are obtained
through specialized texture analysis methods. For classification purpose, selfadvising
SVM is adapted which showed improved classification rate as compared
to standard SVM. The diagnostic accuracy obtained through the proposed
system is around 90% with sensitivity 91% and specificity 89%.

Melanoma is the most deathful form of skin cancer but early diagnosis can ensure a high rate of survival. Early diagnosis is one of the greatest challenges due to lack of experience of general practitioners (GPs). This paper presents a clinical decision support system designed for the use of general practitioners, aiming to save time and resources in the diagnostic process. Segmentation, pattern recognition, and lesion detection are the important steps in the proposed decision support system. The system analyses the images to extract the affected area using a novel proposed segmentation method. It determinates the underlying features which indicate the difference between melanoma and benign images and makes a decision. Considering the efficiency of neural networks in classification of complex data, scaled conjugate gradient based neural network is used for classification. The presented work also considers analyzed performance of other efficient neural network training algorithms on the specific skin lesion diagnostic problem and discussed the corresponding findings. The best diagnostic rates obtained through the proposed decision support system are around 92%.

Melanoma is the most deathly of all skin cancers but early diagnosis can ensure a high degree of survival. Early diagnosis is one of the greatest challenges due to lack of experience of general practitioners (GPs). In this paper we present a clinical decision support system designed for general practitioners, aimed at saving time and resources in the diagnostic process. Segmentation, pattern recognition, and change detection are the important steps in our approach. This paper also investigates the performance of Artificial Neural Network (ANN) learning algorithms for skin cancer diagnosis. The capabilities of three learning algorithms i.e. Levenberg-Marquardt (LM), Resilient Back propagation (RP), Scaled Conjugate Gradient (SCG) algorithms in differentiating melanoma and benign lesions are studied and their performances are compared. The results show that Levenberg-Marquardt algorithm was quick and efficient in figuring out benign lesions with specificity 95.1%, while SCG algorithm gave better results in detecting melanoma at the cost of more number of epochs with sensitivity 92.6%. © 2014 Springer International Publishing Switzerland.

This paper proposes an automated non-invasive system for skin cancer (melanoma) detection based on Support Vector Machine classification. The proposed system uses a number of features extracted from the Wavelet or the Curvelet decomposition of the grayscale skin lesion images and color features obtained from the original color images. The dataset used include both digital images and Dermoscopy images for skin lesions that are either benign or malignant. The recognition accuracy obtained by the Support Vector Machine classifier used in this experiment is 87.7.1% for the Wavelet based features and 83.6. 6% for the Curvelet based ones. The proposed system also resulted in a sensitivity of 86.4 % for the case of Wavelet and
76.9% for the case of Curvelet. It also resulted in a specificity of 88.1% for the case of Wavelet and 85.4% for the case of Curvelet. The obtained sensitivity and specificity results are comparable to those obtained by Dermatologists.

This paper proposes an automated system for discrimination between melanocytic nevi and malignant melanoma. The proposed system used a number of features
extracted from histo-pathological images of skin lesions through image processing techniques which consisted of a spatially adaptive color median lter for ltering and a Kmeans clustering for segmentation. The extracted features were reduced by using sequential feature selection and then classied by using support vector machine (SVM) to diagnose skin biopsies from patients as either malignant melanoma or benign nevi. The proposed system was able to achieve a good result with classication accuracy of 88.9%, sensitivity of 87.5% and specicity of 100%.

The use of a small number of Electromyography (EMG) channels for classifying the finger movement is a challenging task. This paper proposes the recognition system for decoding the individual and combined finger movements using two channels surface EMG. The proposed system utilizes Spectral Regression Discriminant Analysis (SRDA) for dimensionality reduction, Extreme Learning Machine (ELM) for classification and the majority vote for the classification smoothness. The experimental results show that the proposed system was able to classify ten classes of individual and combined finger movements, offline and online with accuracy 97.96 % and 97.07% respectively.

This paper proposes the pattern recognition system for individual and combined finger movements by using two channel electromyography (EMG) signals. The proposed system employs Spectral Regression Discriminant Analysis (SRDA) for dimensionality reduction, Extreme Learning Machine (ELM) for classification and the majority vote for the classification smoothness. The advantage of the SRDA is its speed which is faster than original LDA so that it could deal with multiple features. In addition, the use of ELM which is fast and has similar classification performance to well-known SVM empowers the classification system. The experimental results show that the proposed system was able to recognize the individual and combined fingers movements with up to 98 % classification accuracy by using only just two EMG channels.

Taking the advantage of human brain plasticity nature, Augmented Reality (AR) based Illusion System (ARIS) for upper limb rehabilitation has developed. The ARIS aims to restore the lost functions of upper limb due to various motor injuries. It incorporates with AR technology to build up the upper limb rehabilitation exercise and computer vision with color recognition technique to comply Fool-the-Brain concept for fast recovery of neural impairments. The upper limb exercise that developed in ARIS is to promote the impaired arm range of motion by moving along the predefined trajectory of the AR based exercise. In ARIS, the real impaired arm will be overlapped by the virtual arm throughout the rehabilitation exercise to create the illusion scene. In the case of real arm cannot perform the required task, virtual arm will take over the job of real one and will let the user to perceive the sense that he/she is still able to perform the reaching movement by own effort to the destination point which is the main idea of ARIS. The validation of ARIS was conducted as a preliminary stage and the outcome are discussed.

Sleep apnea (SA) is the most important and common component of sleep disorders which has several short term and long term side effects on health. There are several studies on automated SA detection but not too much works have been done on SA prediction. This paper discusses the application of artificial neural net-works (ANNs) to predict sleep apnea. Three types of neural networks were investigated: Elman, cascade-forward and feed-forward back propagation. We assessed the performance of the models using the Receiver Operating Characteristic (ROC) curve, particularly the area under the ROC curves (AUC), and statistically compare the cross validated estimate of the AUC of different models. Based on the obtained results, generally cascade-forward model results are better with average of AUC around 80%

Segmentation of skin lesion is an important step in the overall automated diagnostic systems used for early detection of skin cancer. Skin lesions can have various different forms which makes segmentation a difficult and complex task. Different methods are present in literature for improving results for skin lesion segmentation. Each method has some pros and cons and it is observed that none of them can be regarded as a generalized method working for all types of skin lesions. The paper proposes an algorithm that combines the advantages of clustering, thresholding and active contour methods currently being used independently for segmentation purposes. A modified algorithm for thresholding based on fusion of Fuzzy C mean clustering and histogram thresholding is applied to initialize level set automatically and also for estimating controlling parameters for level set evolution. The performance of level set segmentation is subject to appropriate initialization, so the proposed algorithm is being compared with some other state-of-the-art initialization methods. The work has been tested on clinical database of 270 images. Parameters for performance evaluation are presented in detail. Increased true detection rate and reduced false positive and false negative errors confirm the effectiveness of the proposed method for skin cancer detection.

Accurate segmentation of skin lesion can play a vital role in early detection of skin cancer. Taking the complexity and varieties of skin lesion images into consideration, we propose a new algorithm that combines the advantages of clustering, thresholding and active contour methods currently being used independently for segmentation purposes. A modified Fuzzy C mean thresholding algorithm is applied to initialize level set automatically and also for estimating controlling parameters for level set evolution. The performance of level set segmentation is subject to appropriate initialization, so the proposed initialization method is compared to some other state of the art initialization methods present in literature. The work has been tested on a clinical database of 238 images. Parameters for performance evaluation are presented in detail. Increased true detection rate and reduced false positive and false negative errors confirm the effectiveness of the proposed method for skin cancer detection.

More than 62,000 Australians were reported suffered from Traumatic Brain Injury (TBI), Spinal Cord Injury (SCI) and Cerebrovascular Accident (CVA) or stroke in 2011. These injuries and accidents lead to physical disability that yields in limitation for performing a personâs daily life activities. To overcome such limitations, physical rehabilitation is conducted which requires one-to-one attention that creates a shortage in therapists and lead to high cost. In this paper, a development of an effective augmented reality (AR) based upper limb rehabilitation system with low cost is presented. Our development aims to close the gap in shortage of therapists, high health care cost of TBI, SCI and stroke. The proposed system can be used at home as well as in rehabilitation centers, units, and hospitals with minimum therapist supervision. It consists of two modules: AR based rehabilitation exercises module and real-time active muscle module. The first module aims to increase the upper limb range of motion via reaching exercises, and strengthen the associated muscles. In second module, the patientâs EMG signals were used as an input to monitor the muscle performance in real time during training. Our development had tested with 10 healthy subjects and had demonstrated in Port Kembla Rehabilitation Hospital.

More than 62,000 Australians were suffered from Traumatic Brain Injury, Spinal Cord Injury and stroke in 2011. The results of such injuries lead to physical disabilities that yield to prohibiting from performing a personâs daily life activities and reduce the quality of life. Moreover, the cost of such affects is almost AUD 13 billion per annum in health care sector. To overcome such situation, rehabilitation is essential for recovery to return to normal life. This paper presents the development of an effective shoulder rehabilitation system with motivational approach. Our development aims to reduce the one-to-one patient-therapist treatment relation contact, make it less regular consultation, and reduce high health care cost. The system is made up of two modules: rehabilitation exercises module where rehabilitation exercises are developed and real-time biofeedback simulation module that detect the active muscle in real time based on sEMG threshold which is predefined by therapist. Four rehabilitation exercises are developed and integrated with biofeedback system. The effectiveness of proposed system has been evaluated through testing with ten subjects and got high performance. The system has demonstrated in Port Kembla Rehabilitation Hospital

In Australia, about 88% of stroke survivors live at home with disabilities affecting their daily life activities and quality of their lives. Therefore, there is a need to improve their lost functions and promote their lives via rehabilitation process. One way to improve the stroke rehabilitation process is through human interactive system, which can be achieved by augmented reality technology. This development draws from the work currently being pursued in the gaming industry to make the augmented reality technology more accessible to the medical industry for the improvement of stroke rehabilitation. In this paper, two augmented reality games: Pong Game and Goal Keeper Game were developed. These games have been designed for rehabilitation with consideration to human interactive systems and have features such as on-screen feedbacks and high immersive value to keep stroke victims motivated in the rehabilitation process. The developed games were aimed to replace boring and repetitive traditional rehabilitation exercises. This paper details the success of implementing augmented reality into the rehabilitation process, which will in turn contribute to society by minimizing the number of people living at home with stroke related disabilities and the requirement for direct supervision from therapist.

Sleep disorders are common in a general population. It effect one in 5 adults and has several short term and long term bad side effects on health. Sleep apnea (SA) is the most important and common component of sleep disorders. This paper presents an automatic approach for detecting apnea events by using few bio-singles that are related to breathe defect. This work uses only air flow, thoracic and abdominal respiratory movement as input. The proposed algorithm consists of three main parts which are signal segmentation, feature generation and classification. A new proposed segmentation method intelligently segments the input signals for further classification, then features are generated for each segment by wavelet packet coefficients and also original signals. In classification phase a unique parallel PSO-SVM algorithm is investigated. PSO used to tune SVM parameters, and also data reduction. Proposed parallel structure used to help PSO to search space more efficiently, also avoiding fast convergence and local optimal results that are common problem in similar parallel algorithms. Obtained results demonstrate that the proposed method is effective and robust in sleep apnea detection and statistical tests on the results shown superiority of it versus previous methods even with more input signals, and also versus single PSO-SVM. Using fewer signals means more comfortable to subject and also, reduction of cost during recording the data.

This paper presents a method for subject independent classification of sleep apnea by a parallel PSO-SVM algorithm. In the proposed structure, swarms are separated into masters and slaves and accessing to the global information is restricted according to their types. Biosignal records that used as the input of the system are air flow, thoracic and abdominal respiratory movement signals. The classification method consists of the three main parts; feature generation, feature selection and data reduction based on parallel PSO-SVM, and the final classification. Statistical analyses on the achieved results show efficiency of the proposed system.

This paper presents a method for signals and features selection when classifying sleep apnea. This study uses a novel hierarchical parallel particle swarm optimization structure as proposed by the authors previously. In this structure, the swarms are separated into `masters and `slaves and access to global information is restricted according to their types. This method is used to classify sleep apneic events into apnea or hypopnea. In this study, ten different biosignals are used as the inputs for the system albeit with different features. The most important signals are subsequently determined based on their contribution to classification of the sleep apneic events. The classification method consists of three main parts which are: feature generation, signal selection, and data reduction based on PSO-SVM, and the final classifier. This study can be useful for selecting the best subset of input signals for classification of sleep apneic events, by attention to the trade of between more accuracy of higher number of input signals and more comfortable of less signals for the patient

In this paper Self-Advising SVM, a new proposed version of SVM, is investigated for sleep apnea classification. Self-Advising SVM tries to transfer more information from training phase to the test phase in compare to the traditional SVM. In this paper Sleep apnea events are classified to central, obstructive or mixed, by using just three signals, airflow, abdominal and thoracic movement, as inputs. Statistical tests show that self-advising SVM performs better than traditional SVM in sleep apnea classification

A gradient vector flow (GVF) snake is proposed in this paper for the segmentation of skin cancer images. In order to make the snake insensitive to noise and be able to remove the hairs, an Adaptive Filter (Wiener and Median filters) is proposed. After the noise and hairs are removed, GVF snake will be used to segment the skin cancer region. The GVF snake extends the single direction and allows it to still be able to track the boundary of the skin cancer even if there are other objects near the skin cancer region. We have proposed new operators to find better edge map in a restored grey scale image. Subjective method has been used by comparing the performance of the proposed gradient vector flow (GVF) snake with other recommended operators of first derivative like Sobel, Prewitt, Roberts and second derivative like Laplacian. The root mean square error and root mean square of signal to noise ratio have been used for objective evaluation. Finally, to validate the efficiency of the filtering schemes different algorithms are proposed and the simulation study has been carried out. Experiments performed on 8(eight) cancer images show the effectiveness of the proposed algorithm.

this paper proposes an automatic skin cancer (melanoma) classification system. The input for the prosed system is a collected data images, it followed by different image processing procedures to enhance the image properties. Two segmentation methods used to identify the normal skin cancer from malignant skin and to extract the useful information from these images that passed to the classifier for training and testing. The features used for classification is the coefficients created by Wavelet decompositions and simple wrapper curvelet. Curvelet is suitable for the image that contains oriented texture and cartoon edges. Recognition accuracy of the three layers back-propagation neural network classifier with wavelet is 51.1% and with curvelet is 75. 6% in digital images database.

Stroke or cerebrovascular accident (CVA) causes disability and affected the personâs quality of life. The rehabilitation therapies are normally conducted for post stroke patients to promote their quality of life and daily living standard. Among rehabilitation exercises, shoulder range of motion (ROM) exercise and muscle strengthening exercise are the most important rehabilitation therapies for post stroke patients as this can improve their activities of daily life. Among the shoulder ROM exercises, the reaching exercise is normally conducted with checkerboard in rehabilitation centre as a traditional therapy which becomes boring after trained for few times. To overcome this problem, same exercise with augmented reality (AR) based game like style incorporate with motivated visual and audio feedbacks has developed and details of the system is presented in this paper. The AR based reaching exercise has developed within the normal average range of motion. The system includes personal computer or laptop, webcam, marker and BioGraph Infiniti system. Thus, it can be used at home without modifying anything and patient can use easily by himself without extra help. The developed system has integrated with biofeedback system to become more effective in rehabilitation. The integrated system has already tested with healthy subject and worked perfectly with positive feedbacks.

Stroke is one of the leading causes of disability in all over the world. This disability greatly impacts the stroke patientsâ daily life activities. Thus, rehabilitation exercises are essential for post stroke patients to restore their lost functions gradually for daily life activities. Traditional rehabilitation exercises do not motivate the post stroke patients as they are normally humdrum and required expensive equipments. Therefore, this paper presents the development of low âcost motivating webcam colour based visual tracking augmented reality (AR) system with biofeedback for upper-limb post stroke rehabilitation therapy. Augmented Reality is a novel form of human-computer interface which overlay the computer-generated information on the real world environment rather than replaces it. In the developed AR system, two games; Ping Pong Rehab (PPR) and Balloon Collection Rehab (BCR) are created based on game design principle. PPR game trains shoulder and arm muscles during rehabilitation therapy whilst BCR game trains shoulder, arm and forearm muscles. Both games have been built and integrated with Biograph Infiniti software to monitor the musclesâ performance. The integrated system will obtain the biofeedback EMG signals from patients that will be utilised for future developments. It allows the patients to monitor their arms and muscles movements in real time on the display screen via low-cost webcam. The system aims for home based rehabilitation system and friendly used by patients themselves. The developed integrated system has tested with able subject and it worked perfectly during the test.

Traumatic Brain Injury (TBI), Spinal Cord Injury (SCI) and Stroke or Cerebrovascular Accident (CVA) cause severe physical disability and affect the person quality of life. Therefore, rehabilitation therapies are essential for those patients to promote their quality of life and restore their lost functions to perform daily live activities. Daily care cost is very high for TBI, SCI and CVA that become major problem for the patients and their families. Moreover, shortage of therapists is one of the major problems in rehabilitation hospitals due to one to one basic training. To overcome these problems, this paper presents low cost motivated rehabilitation system with minimum supervision of therapist for upper limb system. The proposed system can be used as a home based or rehabilitation center therapy system. It is has two modules namely rehabilitation exercise module and real-time muscle simulation module. Several Augmented Reality (AR) games have developed as rehabilitation exercises and integrated with real-time muscle simulation to complete the system. Real-time muscle simulation was achieved based on patient electromyography (EMG) signals in real time. While the system will work to retrain the elastic brain via fast recovery method, it is also will close the gap for the required information, by therapists, about monitoring and tracks the patient muscle performance. The system has tested with five healthy subjects and revealed with potential rehabilitation system for disabled people.

Stroke or cerebrovascular accident (CVA) causes disability and affected the person's quality of life. The rehabilitation therapies are normally conducted for post stroke patients to promote their quality of life and daily living standard. Among rehabilitation exercises, shoulder range of motion (ROM) exercise and muscle strengthening exercise are the most important rehabilitation therapies for post stroke patients as this can improve their activities of daily life. Among the shoulder ROM exercises, the reaching exercise is normally conducted with checkerboard in rehabilitation centre as a traditional therapy which becomes boring after trained for few times. To overcome this problem, same exercise with augmented reality (AR) based game like style incorporate with motivated visual and audio feedbacks has developed and details of the system is presented in this paper. The AR based reaching exercise has developed within the normal average range of motion. The system includes personal computer or laptop, webcam, marker and BioGraph Infiniti system. Thus, it can be used at home without modifying anything and patient can use easily by himself without extra help. The developed system has integrated with biofeedback system to become more effective in rehabilitation. The integrated system has already tested with healthy subject and worked perfectly with positive feedbacks.

Skin cancer is increasing in different countries especially in Australia. Early detection of skin cancer can treat melanoma successfully, therefore, curability and survival depends directly on removing melanoma in its early stages. Since clinical observations face to different fault for melanoma detection, the automatic diagnosis can help to increase the accuracy of detection. Reviewing the researches have done in skin cancer detection and providing the overview on automatic detection of skin cancer are the ultimate aims of this paper. It presents the literature on automatic skin cancer detection and describes the different steps of such process.

It has been reported that 53,000 stroke events annually with ongoing costs are nearly $500 million per year for physical therapy care. This paper aims to provide effective and active rehabilitation for patients suffering from upper limb that a slight or partial paralysis, using gaming based a therapy technique. By disguising the tasks into more entertaining, patients are motivated to train for longer and more frequently. The advantage of this system can be a self-managed, at-home therapy system; reducing fatigue for physical therapists, and the time required for therapistpatient sessions. The system incorporates a virtual reality (VR) environment displaying both the games and a human model as feedback of the patientsâ actions whilst playing the games. Two games were developed; Whack-a-Mouse, and Rolly games, each targeting improvement of muscle strength, control, accuracy and speed. The difficulty of the games can be varied to suit a number of impairments and patient progress is monitored. The games are played using a Nintendo Wii controller. The successful improvements with lower costs associated with this system, are marked improvements for patients suffering from such a debilitating condition.

It has been reported that 53,000 stroke events annually with ongoing costs are nearly $500 million per year for physical therapy care. This paper aims to provide effective and active rehabilitation for patients suffering from upper limb paresis, using gaming based a therapy technique. By disguising the tasks into more entertaining, patients are motivated to train for longer and more frequently. The advantage of this system can be a self-managed, at-home therapy system; reducing fatigue for physical therapists, and the time required for therapist-patient sessions. The system incorporates a virtual reality (VR) environment displaying both the games and a human model as feedback of the patientsâ actions whilst playing the games. Two games were developed, each targeting improvement of muscle strength, control, accuracy and speed. The difficulty of the games can be varied to suit a number of impairments and patient progress is monitored. The games are played using a Nintendo Wii controller. The successful improvements with lower costs associated with this system, are marked improvements for patients suffering from such a debilitating condition.

Sleep Apnea (SA) is one of the common symptoms and important part of sleep disorders. It has consequences that affect all daily life activities and present danger to the patient and/or others. The common diagnose procedure is based on an overnight sleep test. The test is usually including recording of serveral bio-signals that used to detect this syndrome. The conventional apprach of detecting the sleep apnea uses a manual analysis of most of bio-signals to achieve reasonable accuracy. The manual prcess is highly cost and time consuming. This paper presents a novel automatic system for detecting Apnea events by using just few of bio-signals that are related to breathe defect. This work use only, Air flow, thoracic and abdominal respiratory movement as inputs for the system. The proposed technique consist of three main parts which are signal segmnetation, feature generation and classification based on genetic SVM. Results show efficiency of this system as its superiority versus previous methods with more bio-signals as input.

utilizing CCTV cameras based on fuzzy logic model. The propose system aiming to has a capability of identifying certain objects in line of sight and determine if the object is a human or not. The propose system is a real time system for human detection, tracking, and verification in such challenging environments. The system integrates several computer vision techniques in order to design an algorithm that is capable of processing the digital images to perform its functions that include human detection, object tracking, and motion analysis. The testing results provide very encouraging outcome for the proposed system.

Biometric recognition system refers to the automatic recognition of human beings based on their physiological and/or behavioral characteristics. First, a general Automatic Face Recognition (AFR) system framework is proposed to illustrate the processing stages of face images. The AFR includes two main phases: the enrolment phase and recognition/verification phase. Secondly, the main methods of human face recognition are presented, described and compared. Different popular and modern algorithms that have been applied on living persons to verify or recognize the identity based on his/her physiological characteristic are reviewed in this paper. Also, recent face dataset is tabulated to assist the researchers how to find proper database in their applications.

Healthcare is a complex system, so effective technologies and algorithms required to obtaining optimal benefits and results. Also, healthcare has a very large domain but as an engineer we can see that many common algorithms are used for healthcare engineering today's. In this study, we want to review soft computing approaches used for analyze biosignals in healthcare technology issues and specially assistance devices. By attention to variety of assistance devices we can't mention to all published papers in this area. Actually, we will try to introduce some significant algorithms by order of time to provide a good perspective of this area. It must be noted that we don't select papers only based on the number of their citation. But also, some significant papers that are not mention too much, for any reason, are introduced.

The users of current prosthetic and rehabilitation devices are facing problems to adapt to their new hosts or not receiving any bio-feedback despite rehabilitation process and retraining, particularly when working with Electromyogram (EMG) signals. In characterizing virtual human limbs, as a potential prosthetic device in 3D virtual reality, patients are able to familiarize themselves with their new appendage and its capabilities in a virtual training environment or can see their movements intention. This paper presents a Virtual Reality (VR) based design and implementation of a below-shoulder 3D human arm capable of 10-class EMG based motions driven system of biomedical EMG signal. The method considers a signal classification output as potential control stimulus to drive the virtual prosthetic prototype. A hierarchical design methodology is adopted based on anatomical structure, congruent with Virtual Reality Modeling Language (VRML) architecture. The resulting simulation is based on a portable, self-contained VR model implementation paired with an instrumental virtual control-select board capable of actuating any combinations of singular or paired kinematic 10-class EMG motions. The built model allows for multiple degree of freedom profiles as the classes can be activated independently or in conjunction with others allowing enhanced arm movement.

The users of current prosthetic and rehabilitation devices are facing problems to adapt to their new hosts or not receiving any bio-feedback despite rehabilitation process and retraining, particularly when working with Electromyogram (EMG) signals. In characterizing virtual human limbs, as a potential prosthetic device in 3D virtual reality, patients are able to familiarize themselves with their new appendage and its capabilities in a virtual training environment or can see their movements' intention. This paper presents a Virtual Reality (VR) based design and implementation of a below-shoulder 3D human arm capable of 10-class EMG based motions driven system of biomedical EMG signal. The method considers a signal classification output as potential control stimulus to drive the virtual prosthetic prototype. A hierarchical design methodology is adopted based on anatomical structure, congruent with Virtual Reality Modeling Language (VRML) architecture. The resulting simulation is based on a portable, self-contained VR model implementation paired with an instrumental virtual control-select board capable of actuating any combinations of singular or paired kinematic 10-class EMG motions. The built model allows for multiple degree of freedom profiles as the classes can be activated independently or in conjunction with others allowing enhanced arm movement. © 2010 ACM.

Computer vision techniques have been widely used in various applications. In recent years, as energy efficiency have gradually become a important issues, computer vision techniques can be integrated into a smart control system that helps increase the energy efficiency by controlling the turn on of the light based on human detection . However, implement such system that detect walking human in a semi-dark environment remains a challenge. This paper proposes a novel detection technique combining movement analysis and SVM classifier to tackle this problem. This technique consist of a few steps: a statistical background model to segment moving objects as foreground, followed by an analysis model to generate pedestrian candidates based on the movement of foreground objects and lastly a SVM classifier that verify the pedestrian candidates based on the shape features.

In this paper, an automatically skin cancer classification system is developed and the relationship of skin cancer image across different type of neural network are studied with different types of preprocessing.. The collected images are feed into the system, and across different image processing procedure to enhance the image properties. Then the normal skin is removed from the skin affected area and the cancer cell is left in the image. Useful information can be extracted from these images and pass to the classification system for training and testing. Recognition accuracy of the 3-layers back-propagation neural network classifier is 89.9% and auto-associative neural network is 80.8% in the image database that include dermoscopy photo and digital photo.

Large scale wireless sensor networks present interesting challenges in many applications due to a number of factors, such as, energy constraints, the dynamic nature of the network topology and various application needs. An important issue is how the sensor nodes can achieve efficient global behavior under distributed control mechanisms? One of many possible solutions is to construct a self-organized wireless sensor network to deal with this challenge. This paper presents an algorithm to construct a self-organized wireless sensor network based on two dimensional cellular automata that can provide better understanding for how local behavior at node level influences the overall system behavior and affect the system performance. Two types of Cellular Automata (CA) are considered; for the synchronous CA based system; the regular patterns are identified and discussed. To overcome some limitations arising from the use of a synchronous CA implementation an asynchronous CA is employed.

Feature selection is an indispensable pre-processing step when mining huge datasets that can significantly improve the overall system performance. This paper presents a novel feature selection method that utilizes both the Ant Colony Optimization (ACO) and fuzzy memberships. The algorithm estimates the local importance of subsets of features, i.e., their pheromone intensities by utilizing fuzzy c-means (FCM) clustering technique. In order to prove the effectiveness of the proposed method, a comparison with another powerful ACO based feature selection algorithm that utilizes the Mutual Information (MI) concept is presented. The method is tested on two biosignals driven applications: Brain Computer Interface (BCI), and prosthetic devices control with myoelectric signals (MES). A linear discriminant analysis (LDA) classifier is used to measure the performance of the selected subsets in both applications. Practical experiments prove that the new algorithm can be as accurate as the original method with MI, but with a significant reduction in computational cost, especially when dealing with huge datasets.

In this paper, a novel feature selection algorithm based on differential evolution (DE) optimization technique is presented. The new algorithm, called DEFS, modifies the DE which is a real-valued optimizer, to suit the problem of feature selection. The proposed DEFS highly reduces the computational costs while at the same time proving to present powerful performance. The DEFS technique is applied to a brain-computer-interface (BCI) application and compared with other dimensionality reduction techniques. The practical results indicate the significance of the proposed algorithm in terms of solutions optimality, memory requirement, and computational cost.

The myoelectric signal (MES) from human muscles is usually utilized as an input to the controller of a multifunction prosthetic hand. In such a system, a pattern recognition approach is usually employed to discriminate between the MES from different classes. Since the MES is recorded using multi channels, the feature vector size can become very large. In order to reduce the computational cost and enhance the generalization capability of the classifier, a dimensionality reduction method is needed to identify an informative moderate size feature set. This paper proposes a novel feature projection technique based on a combination of Fisher's Linear Discriminant Analysis (LDA), and Fuzzy Logic. The new method, called FLDA, assigns different membership degrees to the data points thus reducing the effect of overlapping points in the discrimination process. Furthermore, the concept of Mutual Information (MI) is introduced in the fuzzy memberships in order to assign weights to the features (attributes) according to their contribution to the discrimination process. The FLDA method is tested on a seven classes MES dataset and compared with other feature projection techniques proving its superiority.

The myoelectric signals (MES) from human muscles have been utilized in many applications such as prosthesis control. The identification of various MES temporal structures is used to control the movement of prosthetic devices by utilizing a pattern recognition approach. Recent advances in this field have shown that there are a number of factors limiting the clinical availability of such systems. Several control strategies have been proposed but deficiencies still exist with most of those strategies especially with the Dimensionality Reduction (DR) part. This paper proposes using Particle Swarm Optimization (PSO) algorithm with the concept of Mutual Information (MI) to produce a novel hybrid feature selection algorithm. The new algorithm, called PSOMIFS, is utilized as a DR tool in myoelectric control problems. The PSOMIFS will be compared with other techniques to prove the effectiveness of the proposed method. Accurate results are acquired using only a small subset of the original feature set producing a classification accuracy of 99% across a problem of ten classes based on tests done on six subjects MES datasets.

The myoelectric signals (MES) from human muscles have been utilized in many applications such as prosthesis control. The identification of various MES temporal structures is used to control the movement of prosthetic devices by utilizing a pattern recognition approach. Recent advances in this field have shown that there are a number of factors limiting the clinical availability of such systems. Several control strategies have been proposed but deficiencies still exist with most of those strategies especially with the Dimensionality Reduction (DR) part. This paper proposes using Particle Swarm Optimization (PSO) algorithm with the concept of Mutual Information (MI) to produce a novel hybrid feature selection algorithm. The new algorithm, called PSOMIFS, is utilized as a DR tool in myoelectric control problems. The PSOMIFS will be compared with other techniques to prove the effectiveness of the proposed method. Accurate results are acquired using only a small subset of the original feature set producing a classification accuracy of 99% across a problem of ten classes based on tests done on six subjects MES datasets.

Feature selection is an important step in many pattern recognition systems that aims to overcome the so-called curse of dimensionality problem. Although Ant Colony Optimization (ACO) proved to be a powerful technique in different optimization problems, but it still needs some improvements when applied to the feature selection problem. This is due to the fact that it builds its solutions sequentially, where in feature selection this behavior will most likely not lead to the optimal solution. In this paper, a novel feature selection algorithm based on a combination of ACO and a simple, yet powerful, Differential Evolution (DE) operator is presented. The proposed combination enhances both the exploration and exploitation capabilities of the search procedure. The new algorithm is tested on two biosignal-driven applications. The performance of the proposed algorithm is compared with other dimensionality reduction techniques to prove its superiority.

The myoelectric signal (MES) from human muscles is usually utilized as an input to the controller of a multifunction prosthetic hand. In such a system, a pattern recognition approach is usually employed to discriminate between the MES from different classes. Since the MES is recorded using multi channels, the feature vector size can become very large. In order to reduce the computational cost and enhance the generalization capability of the classifier, a dimensionality reduction method is needed to identify an informative moderate size feature set. This paper proposes a novel feature projection technique based on a combination of Fisher's Linear Discriminant Analysis (LDA), and Fuzzy Logic. The new method, called FLDA, assigns different membership degrees to the data points thus reducing the effect of overlapping points in the discrimination process. Furthermore, the concept of Mutual Information (MI) is introduced in the fuzzy memberships in order to assign weights to the features (attributes) according to their contribution to the discrimination process. The FLDA method is tested on a seven classes MES dataset and compared with other feature projection techniques proving its superiority. © 2008 IEEE.

n this paper, a new feature extraction method utilizing ant colony optimization in the selection of wavelet packet transform (WPT) best basis is presented and adopted in classifying biomedical signals. The new algorithm, termed intelligent artificial ants (IAA), searches the wavelet packet tree for subsets of features that best interact together to produce high classification accuracies. While traversing the WPT tree, the IAA takes into account existing correlation between features thus avoiding information redundancy. The IAA method is a mixture of filter and wrapper approaches in feature subset selection. The pheromone that the ants lay down is updated by means of an estimation of the information contents of a single feature or feature subset. The significance of the subsets selected by the ants is measured using linear discriminant analysis (LDA) classifier. The IAA method is tested on one of the most important biosignal driven applications, which is the brain computer interface (BCI) problem with 56 EEG channels. Practical results indicate the significance of the proposed method achieving a maximum accuracy of 83%.

Recent approaches in multifunction myoelectrically controlled prosthetic devices revealed that dimensionality reduction plays a significant role in the overall system performance. In this paper, a new feature selection method is developed based on a mixture of particle swarm optimization (PSO) method and the concept of mutual information (MI). The new method, termed PSO-MI, is adopted as a dimensionality reduction tool for myoelectric control. The PSO-MI employs the MI measure to aid in controlling the movements of particles in the solution space, thus forming a kind of a hybrid filter-wrapper method. The new PSO-MI is able to account for the interaction property between the features in the selected subset, thus producing high classification accuracies. A dataset of transient myoelectric signal (MES) consisting of six classes of hand grasp is utilized to test the performance of the proposed method. It is proved that the PSO-MI outperforms other methods adopted for dimensionality reduction in myoelectric control achieving 95.5% of classification accuracy across six classes problem.

In this paper, a novel feature extraction method based on the utilization of wavelet packet transform (WPT) and the concept of fuzzy entropy is presented. The method acts in steps, were in the first step the WPT is employed to generate a wavelet decomposition tree from which many features are extracted. In the second step, a new algorithm to compute the fuzzy entropy is developed and adopted as a measure of information content to judge on features suitability in classification, by setting a threshold and removing the features that fall under a certain threshold. In the third step, principle component analysis (PCA) is employed to reduce the dimensionality of the generated feature set. As an application, the new algorithm is employed in multifunction myoelectric control problem to prove its efficiency. Accurate results (99% accuracy) are acquired from using only a small subset of the original feature set generated by the wavelet tree. The obtained results indicate that the generated features are of maximum relevance and with minimum degree of redundancy.

Many intelligent techniques were established during last decades to handle nonlinear, multimode, noisy, nondifferentiable problems and to obtain optimum solution(s). This paper presents improving and implementations for two recently intelligent techniques; type-2 fuzzy system (T2 FS) and particle swarm optimization (PSO) and presents a new method to optimize parameters of the primary membership functions of T2 FS by PSO to improve the performance and increase the accuracy of T2 FS model. The implementation of the suggested method on mean blood pressure estimation has very successful rate.

In this paper, a new intelligent method of classifying benign and malignant melanoma lesions is implemented. The system consists of four stages; image pre-processing, image segmentation, feature extraction, and image classification. As the first step of the image analysis, pre-processing techniques are implemented to remove noise and undesired structures from the images using techniques such as median filtering and contrast enhancement. In the second step, a simple thresholding method is used to segment and localise the lesion, a boundary tracing algorithm is also implemented to validate the segmentation. Then, a wavelet approach is used to extract the features, more specifically wavelet packet transform (WPT). Finally, the dimensionality of the selected features is reduced with principal component analysis (PCA) and later supplied to an artificial neural network and support vector machine classifiers for classification. The ability to correctly discriminate between benign and malignant lesions was about 95% for the Artificial Neural Network and 85% for the Support Vector Machine classifier.

Real time controlling devices based on myoelectric singles (MES) is one of the challenging research problems. This paper presents a new approach to reduce the computational cost of real time systems driven by Myoelectric signals (MES) (a.k.a Electromyography -EMG). The new approach evaluates the significance of feature/channel selection on MES pattern recognition. Particle Swarm Optimization (PSO), an evolutionary computational technique, is employed to search the feature/channel space for important subsets. These important subsets will be evaluated using a multilayer perceptron trained with back propagation neural network (BPNN). Practical results acquired from tests done on six subject's datasets of MES signals measured in a noninvasive manner using surface electrodes are presented. It is proved that minimum error rates can be achieved by considering the correct combination of features/channels, thus providing a feasible system for practical implementation purpose for rehabilitation of patients.

Real time controlling devices based on myoelectric singles (MES) is one of the challenging research problems. This paper presents a new approach to reduce the computational cost of real time systems driven by Myoelectric signals (MES) (a.k.a Electromyography -EMG). The new approach evaluates the significance of feature/channel selection on MES pattern recognition. Particle Swarm Optimization (PSO), an evolutionary computational technique, is employed to search the feature/channel space for important subsets. These important subsets will be evaluated using a multilayer perceptron trained with back propagation neural network (BPNN). Practical results acquired from tests done on six subject's datasets of MES signals measured in a noninvasive manner using surface electrodes are presented. It is proved that minimum error rates can be achieved by considering the correct combination of features/channels, thus providing a feasible system for practical implementation purpose for rehabilitation of patients. © 2007 IEEE.

Blood pressure measurement is a highly sensitive task, because even breathing can cause variation as high as 10 mmHg in reading of blood pressure. Due to the presence of high level of uncertainty; the linear model for blood pressure classification is not appropriate. Fuzzy logic systems are capable of producing precise solutions from vague, incomplete, or approximate information, by accommodating the ambiguities and logic. This paper presents a novel type-2 Fuzzy Logic System to estimate and classify blood pressure parameters in appropriate linguistic description. Firstly, a type-2 fuzzy logic system for the classification of blood pressure parameters is designed. Secondly, the proposed model is demonstrated by graphical user interface. The designed fuzzy model for the classification of blood pressure parameters gives more realistic results than linear model. The outcome of this paper is a friendly graphic user interface (GUI). The proposed model takes crisp value of heart rate as an input and generates crisp reading of blood pressure along with its appropriate linguistic classification, e.g., normal, low, or high etc; to provide more clear information to the general public about their cardiac health. The system has been tested and the computed percentage is less than 10% error values of all ten subjects' systolic, diastolic and mean blood pressure.

Mean blood pressure (MBP) has high correlation with heart rate (HR), but such e relationship between them is ambiguous and nonlinear. This paper investigates establishing an accurate mathematical model to estimate MBP that is considering the influence of the stroke volume changing. Twenty three cases of MIMIC database till are employed; 12 cases for training and 11 cases for verification. The mean and standard deviation for all cases are calculated and compared with real results. Our suggested mathematical model achieved an encouragement results

The mean arterial pressure is very important parameter for physicians to diagnose various cardiovascular diseases. There are many algorithms, with different accuracy, to estimate arterial blood pressure that used different factors such as blood level, pulses and external applied pressure, photoplethysmography signal features, heart rate and other factors. This paper develops an algorithm; which utilizes the heart rate with considering stroke volume (SV) influence; to estimate the mean arterial pressure. The SV influence is computed by particle swarm optimization technique. This new algorithm has developed relayed on twenty cases of mimic database [1]; ten cases for training and all cases for verification. This algorithm achieved encouragement results and get very close estimated MAP to the real MAP. © 2006 Research Publishing Services.

Many cardiovascular diseases can be avoided by continuous monitoring cardiovascular parameters. Heart rate, electrocardiogram, blood pressure and pulse wave velocity are the most important and popular cardiovascular parameters. These parameters can be measured by different sensors that have been developed and improved to achieve reliable, accurate and continuous measurements. A part of the processing of theses sensors data, to get the related information, is parameters estimation.

Localisation for human or robots in built environment is one of impartment issue in many applications. Increasing mobile computing devices, that often require knowing where things are actually placed, can be utilised and can be used in localisation systems. To use such systems, various and different location systems and technologies have been developed. This paper studies the use of the radio frequency signal strength information extracted from a pre-installed wireless network as a localiser sensor to locate a mobile (human) user in built environment. A strategy based on a fuzzy logic technique to estimate the mobile user location in built environment has been investigated

A multi agent approach and robotics behaviors can be classified as a type of distributed intelligence. This distributed intelligence has an advantage to solving particular problems through benefiting from "divide and conquer" approach. Inherent parallelism of multiple agents can do many, perhaps different things, at the same time. Multiple agents are able to be in different places at the same time and can achieve different goals at the same time. Our paper is presenting achieving of static and dynamic goals completion through agent negotiation. The robotic agents shall assist each other in goal acquisition. This can be achieved in any environment with no map or determined damnation. The agents in our work do not necessarily know it's location in the environment during its goal acquisition. Our work is allowing easy location communication between agents in indoor environment.

This paper describes a multi-agent system that performs exploration and target seeking. A bidding scheme is applied to obtain assistance and cooperation between the mobile robots. An occupancy grid-based map and laser sensors are used for mapping. Frontier based exploration is implemented to determine unexplored areas to search. Targets are identified by designated colours in the vision sensors. Wave front propagation is used for path planning to both the frontiers and the targets. Fuzzy logic is applied to the laser readings to perform collision avoidance.

The paper presents an experimental study of a probabilistic road map (PRM) based obstacle avoiding algorithm, for motion planning of a non-holonomic mobile robot in a cluttered dynamic environment. The PRM approach uses a fast and simple local planner to build a network representation of the configuration space. It trades off the distance to both static objects and moving obstacles in computing the travelled path. Our work has been implemented and tested on Player/Stage, a real time robotic software, in extensive simulation runs. The different experiments demonstrate that our approach is well suited to control the motions of a robot in a cluttered environment and demonstrates its advantages over other techniques.

© 2000 IEEE. The reaction of autonomous mobile robot to the dynamic, uncertain, and changeable environment is one of most difficult issue in the control of intelligent autonomous robot movement. Fuzzy control appears as a very useful tool for handling the intelligent reactive navigation. Present work deals with building of fiizzy? behavior based reactive navigation of goal achieving, obstacle avoidance, and propose new method to blend and coordinate multi behaviors in the same time. This method using fuzzy technique and fix priority value reflects the importance of the behavior We build our simulation and animation programs that can reflect on-line the robot movement and have capability' to using graphical user interface (GUI).

The reaction of autonomous mobile robot to the dynamic, uncertain, and changeable environment is one of most difficult issue in control of the intelligent autonomous robot movement. Fuzzy control appears as a very useful tool for handling the intelligent reactive navigation. Present work deals with building of fuzzy behaviors based reactive navigation to reach the goal, propose new method to blend and coordinate multi behaviors in the same time. This method using fuzzy technique and fix priority value reflects the importance of the behavior. Our behaviors system is test by building our simulation and animation programs that used to reflect on-line the robot movement using graphical user interface (GUI).

The reaction of an autonomous mobile robot to dynamic, uncertain, and changeable environments is one of most difficult issues in control of intelligent autonomous robot movement. Fuzzy control appears as a very useful tool for handling intelligent reactive navigation. The present work deals with building of fuzzy behavior based reactive navigation for obstacle avoidance, and proposes a method to blend and coordinate multi-behaviors at the same time. This method using a fuzzy technique and fixed priority value to reflect the importance of the behavior. We have built our simulation and animation programs that can reflect online the robot movement using a graphical user interface. © 1999 IEEE.