Michael Braun is a medical physicist with particular interest in imaging. He has worked in the areas of image registration, segmentation, tomographic reconstruction, motion correction, tissue characterization with ultrasound, x-ray detector quality, and aspects of magnetic resonance imaging. His current work tends towards assessment of mechanical function based on image data. From time to time, he works on images generated outside the clinical realm, e.g. x-ray CT images of earthworm burrows.
Michael Braun has been a member of a the following professional associations:
- Australasian College of Physical Scientists & Engineers in Medicine
- Institute of Physical Sciences in Medicine (UK)
- Society of Magnetic Resonance in Medicine
- Society for Magnetic Resonance Imaging
- Institute of Electrical & Electronics Engineers
- Royal Australasian College of Radiologists (Assoc.)
- Australian Society for Ultrasound in Medicine
- Australian Institute of Physics
Current interests include:
- analysis of medical images
- assessment of mechanical function of tissues based on clinical images (and, similarly for other materials)
- image registration
- educational resources
- Medical Imaging
- Imaging Science
- Advanced Mechanics
- 1st year physics subjects
- digital signal & image processing
Braun, M, Kirkup, L & Chadwick, S 2018, 'The impact of inquiry orientation and other elements of cultural framework on student engagement in first year laboratory programs', International Journal of Innovation in Science and Mathematics Education, vol. 26, no. 4, pp. 30-48.
© 2018 Institute for Innovation in Science and Mathematics Education. Inquiry-oriented approaches to learning have gradually entered science laboratory programs, aiming to deliver an authentic experience of doing science, enhance student engagement with the material, and bring greater emphasis on generic skills underpinning graduate attributes. Although such approaches have demonstrated pedagogical advantages and improved student engagement, it is not clear how the advantages should be weighted against other elements of what may be regarded as the laboratory program's cultural framework. We analysed two large-enrolment introductory tertiary programs: physics and chemistry at the University of Technology Sydney. The programs differed in the level of inquiry orientation but also in approaches to design, logistics and relevancy. We found that, based on student survey responses, the putative advantages of a deeper inquiry orientation in the physics laboratory were insufficient to compensate for the apparent advantages arising from the other elements of the cultural framework in the chemistry laboratory.
Braun, M 2017, 'Comparative evaluation of online and in-class student team presentations', Journal of University Teaching and Learning Practice, vol. 14, no. 3.
© 2017, University of Wollongong. All rights reserved. Student team presentations are commonly utilised in tertiary science courses to help students develop skills in communication, teamwork and literature research, but they are subject to constraints arising from class size, available time, and limited facilities. In an alternative approach, student teams present online using a variety of tools, such as screencast and blended media, but it is not clear whether this offers an authentic alternative to in-class experience. In this study, the two modes of presentation were compared in terms of student perceptions and academic performance. A survey probed students' familiarity with digital technology, presentation anxiety, and differential perceptions of the two modes. Aside from a confirmation bias, no significant difference was found between those who presented in class and online. In a notable exception, a clear asymmetry appeared when students were asked to choose a mode for a future presentation: none of the online presenters opted for the in-class mode while a third of in-class presenters selected the online mode. Presentation anxiety was similar for in-class and online presenters and was insensitive to gender and familiarity with English. No significant difference was detected between the modes in terms of academic performance.
Braun, M & Kirkup, L 2016, 'Non-physics peer demonstrators in undergraduate laboratories: A study of students' perceptions', European Journal of Physics, vol. 37, no. 1, pp. 1-9.View/Download from: Publisher's site
© 2016 IOP Publishing Ltd. Laboratory demonstrators play a crucial role in facilitating students' learning in physics subjects. Inspired by the success of peer-led activities, we introduced peer demonstrators to support student learning in first-year physics subjects that enrol students not intending to major in physics. Surveys were administered to 1700 students over 4 years in four subjects to examine student perceptions of how demonstrators assisted them in the laboratory. Scores awarded to peer demonstrators by students were no lower than those awarded to demonstrators traditionally employed in the first year physics laboratory. These latter demonstrators were drawn mainly from the ranks of physics research students. The findings validate the recruitment of peer demonstrators and will be used to inform the recruitment and support programmes for laboratory demonstrators.
Kirkup, L, Varadharajan, M & Braun, M 2016, 'A Comparison of Student and Demonstrator Perceptions of Laboratory-Based, Inquiry-Oriented Learning Experiences', International Journal of Innovation in Science and Mathematics Education, vol. 24, no. 2, pp. 1-13.
Paul, J, Prasad, M, Venkatesan, R & Braun, M 2013, 'Magnetic Resonance Scan-time Reduction Using Echo Prediction', International Journal of Imaging Systems and Technology, vol. 23, no. 1, pp. 1-8.View/Download from: Publisher's site
A linear prediction (LP) filter derived from a complete echo with zero-phase encoding amplitude is used for recovering anatomical details from a partially acquired echo sequence. The LP filter is shown to reconstruct missing k-space phase and amplitude i
Yunusa, IA, Braun, M & R, L 2009, 'Amendment of soil with coal fly ash modified burrowing habits of two earthworm species', Applied Soil Ecology, vol. 42, no. 1, pp. 63-68.View/Download from: Publisher's site
A good understanding of how soil biota responds to amendment of agricultural soils with coal fly ash is imperative to developing protocols for routine use of this industrial by-product for soilmanagement. We used X-ray computed tomography (CT) images to determined key structural characteristics of burrows created by earthworms of native megascolecid and exotic Aporrectodea trapezoides in intact soil cores (150 mm ID by 0.3 m deep) that were treated with coal fly ash at 0, 5 or 25 Mg ha1 mixed into the top 50 mm of the cores. The cores were inoculated at a rate equivalent to 850 worms m2 and after 6 weeks we found that fly ash reduced the total volume of the burrow system (Vs) by up to 39% for the native species and 29% for the exotic species due mostly to fewer and smaller burrows; these reductions averaged 33% with addition of ash at 5 Mg ha1 and 39% at 25 Mg ha1. While the native earthworms responded to treatment by burrowing deeper into the soil core and away from the ash-tainted surface soil, the exotic species reduced the depth of burrowing and remained close to the surface. Fly ash addition did not have significant effect on tortuosity (t) of the burrows for either earthworm species. A. trapezoides created predominantly vertical burrows, while the native megascolecid worms produced more horizontally oriented burrows in addition to vertical ones. These modifications of earthworm behavior by fly ash addition to soil, along with previous experience with plant growth, suggest that an ash application rate of 5 Mg ha1is close to optimum for routine agronomic applications. Structural analysis of the burrows as presented in this paper provide more useful information on the response of earthworm behaviour to fly ash that may not be apparent from an assessment of population and growth of these important soil biota.
Hutton, BF, Olsson, A, Erlandsson, K & Braun, M 2006, 'Reducing the influence of spatial resolution to improve quantitative accuracy in emission tomography: a comparison of potential strategies', Nuclear Instrumentation & Methods in Physics Research Section A, vol. 569, no. 2, pp. 462-466.View/Download from: Publisher's site
The goal of this paper is to compare strategies for reducing partial volume effects by either minimising the cause (i.e. improving resolution) or correcting the effect. Correction for resolution loss can be achieved either by modelling the resolution for ise in iterative reconstruction or by imposing contraints based on knowledge of the underlying anatomy. Approaches to partial volume correctiuon largely rely on knoweldge of the underlying anatomy, based on well-registered high-resolution anatomical imaging modalities (CT ot MRI). Corrections can be applied bu considering the signal loss that results by smoothing the high-resolution modality to the same resolution as obntained in emission tomography. A physical phantom representing the central brain structures was used to evaluate the quantitative accuracy of the various strategies for either improving resolution or correcting for partial volume effects. Inclusion of resolution in the reconstruction model imporved the measured contrast for the central brain structures but still underestimated the true object contrast (~0.70). Use of information on the boundaries of the structures in conjunction with a smoothing prior using maximum entropy reconstruction achieved some degree of contrast enhancement and improved the noise properties of the resulting images. Partial volume correction based on segmentation of registered anatomical images and knowledge of the reconstructred resolution permitted more accurate quantificationm of the target to background ration for individual brain structures.
Zhang, Z & Braun, M 2003, 'Smoothness-based forces for deformable models: a long-range force and a corner fitting force', Computers In Biology And Medicine, vol. 33, no. 1, pp. 91-112.View/Download from: Publisher's site
Hutton, BF, Braun, M, Thurfjell, L & Lau, DY 2002, 'Image registration: an essential tool for nuclear medicine', European Journal of Nuclear Medicine, vol. 29, no. 4, pp. 559-577.
Lau, YH, Braun, M & Hutton, BF 2001, 'Non-Rigid Image Registation Using a Median-Filetered Coarse-to-Fine Displacement Field and a Symmetric Correlation Ratio', Physics in Medicine & Biology, vol. 46, no. 2, pp. 1297-1319.View/Download from: Publisher's site
Conventional approaches to image registration are generally limited to image-wide rigid transformations. However, the body and its internal organs are non-rigid structures that change shape due to changes in the body's posture during image acquisition, and due to normal, pathological and treatment-related variations. Inter-subject matching also constitutes a non-rigid registration problem. In this paper, we present a fully automated non-rigid image registration method that maximizes a local voxel-based similarity metric. Overlapping image blocks are defined on a 3D grid. The transformation vector field representing image deformation is found by translating each block so as to maximize the local similarity measure. The resulting sparsely sampled vector field is median filtered and interpolated by a Gaussian function to ensure a locally smooth transformation. A hierarchical strategy is adopted to progressively establish local registration associated with image structures at diminishing scale. Simulation studies were carried out to evaluate the proposed algorithm and to determine the robustness of various voxel-based cost functions. Mutual information, normalized mutual information, correlation ratio (CR) and a new symmetric version of CR were evaluated and compared. A T1-weighted magnetic resonance (MR) image was used to test intra-modality registration. Proton density and T2-weighted MR images of the same subject were used to evaluate inter-modality registration. The proposed algorithm was tested on the 2D MR images distorted by known deformations and 3D images simulating inter-subject distortions. We studied the robustness of cost functions with respect to image sampling. Results indicate that the symmetric CR gives comparable registration to mutual information in intra- and inter-modality tasks at full sampling and is superior to mutual information in registering sparsely sampled images.
Som, S, Hutton, B & Braun, M 2001, 'Optimising use of adaptive filtering and anatomical prior for SPECT reconstruction of clinical data', ANZ Nuclear Medicine, vol. 32, no. 3, pp. 99-104.
This paper reports results demonstrated from phantom and clinical studies, showing that the iterative minimum cross-entropy (MXE) technique, using adaptive filtering and anatomical information, improves the quality (noise and contrast) of SPECT reconstruction compared with maximum likelihood-expectation maximisation (ML-EM) technique. Suitable parameters for using adaptive filtering in the MXE reconstruction of clinical data were first determined. This demonstrated that combined anatomical and physiological information is acceptable for adaptive filtering in MXE reconstruction of clinical SPECT data. Using data obtained from brain images of 15 healthy volunteers, the two techniques are compared by statistical parametric mapping (SPM) analysis. SPM produces significant differences (p<0.001) in ventricle region, which correlates well with findings in region of interest (ROI) analysis.
Fulton, RR, Eberl, S, Meikle, SR, Button, BF & Braun, M 1999, 'A practical 3d tomographie method for correcting patient head motion in clinical SPECT', IEEE Transactions on Nuclear Science, vol. 46, no. 3 PART 2, pp. 667-672.View/Download from: Publisher's site
Patient motion during brain SPECT studies can degrade resolution and introduce distortion. We have developed a correction method which incorporates a motion tracking system to monitor the position and orientation of the patient's head during acquisition. Correction is achieved by spatially repositioning projections according to measured head movements and reconstructing these projections with a fully three-dimensional (3D) algorithm. The method has been evaluated in SPECT studies of the Hoffman 3D brain phantom performed on a triple head camera with fan beam collimation. Movements were applied to the phantom and recorded by a head tracker during SPECT acquisition. Fully 3D reconstruction was performed using the motion data provided by the tracker. Correction accuracy was assessed by comparing the corrected and uncorrected studies with a motion free study, visually and by calculating mean squared error (MSB). In all studies, motion correction reduced distortion and improved MSB by a factor of 2 or more. We conclude that this method can compensate for head motion under clinical SPECT imaging conditions. © 1999 IEEE.
Fulton, RR, Eberl, S, Meikle, SR, Hutton, BF & Braun, M 1999, 'Practical 3D tomographic method for correcting patient head motion in clinical SPECT', IEEE Nuclear Science Symposium and Medical Imaging Conference, vol. 3, pp. 1770-1775.
Patient motion during brain SPECT studies can degrade resolution and introduce distortion. We have developed a correction method which incorporates a motion tracking system to monitor the position and orientation of the patient's head during acquisition. Correction is achieved by spatially repositioning projections according to measured head movements and reconstructing these projections with a fully three-dimensional (3D) algorithm. The method has been evaluated in SPECT studies of the Hoffman 3D brain phantom performed on a triple head camera with fan beam collimation. Movements were applied to the phantom and recorded by a head tracker during SPECT acquisition. Fully 3D reconstruction was performed using the motion data provided by the tracker. Correction accuracy was assessed by comparing the corrected and uncorrected studies with a motion free study, visually and by calculating mean squared error (MSE). In all studies, motion correction reduced distortion and improved MSE by a factor of 2 or more. We conclude that this method can compensate for head motion under clinical SPECT imaging conditions.
Ardekani, BA, Kershaw, J, Braun, M & Kanno, I 1997, 'Automatic detection of the mid-sagittal plane in 3-D brain images', IEEE Transactions on Medical Imaging, vol. 16, no. 6, pp. 947-952.View/Download from: Publisher's site
This article presents a detailed description of an algorithm for the automatic detection of the mid-sagittal plane in three-dimensional (3-D) brain images. The algorithm seeks the plane with respect to which the image exhibits maximum symmetry. For a given plane, symmetry is measured by the cross-correlation between the image sections lying on either side. The search for the plane of maximum symmetry is performed by using a multiresolution approach which substantially decreases computational time. The choice of the starting plane was found to be an important issue in optimization. A method for selecting the initial plane is presented. The algorithm has been tested on brain images from various imaging modalities in both humans and animals. Results were evaluated by visual inspection by neuroradiologists and were judged to be consistently correct. © 1997 IEEE.
Poon, CS & Braun, M 1997, 'Image segmentation by a deformable contour model incorporating region analysis', Physics in Medicine and Biology, vol. 42, no. 9, pp. 1833-1841.View/Download from: Publisher's site
Deformable contour models are useful tools for image segmentation. However, many models depend mainly on local edge-based image features to guide the convergence of the contour. This makes the models sensitive to noise and the initial estimate. Our model incorporates region-based image features to improve its convergence and to reduce its dependence on initial estimation. Computational efficiency is achieved by an optimization strategy, modified from the greedy algorithm of Williams and Shah. The model allows a simultaneous optimization of multiple contours, making it useful for a large variety of segmentation problems.
Ardekani, BA, Braun, M, Hutton, BF, Kannof, I & Iida, H 1996, 'Minimum cross-entropy reconstruction of PET images using prior anatomical information', Physics in Medicine and Biology, vol. 41, no. 11, pp. 2497-2517.View/Download from: Publisher's site
An algorithm is presented for the reconstruction of PET images using prior anatomical information derived from MR images of the same subject. The cross-entropy or Kullback-Leiber distance is a measure of dissimilarity between two images. We propose to reconstruct PET images by minimizing a weighted sum of two cross-entropy terms. The first is the cross-entropy between the measured emission data and the forward projection of the current estimate of the PET image. Minimizing this term alone is equivalent to the ML-EM reconstruction. The second term is the cross-entropy between the current estimate of the PET image and a prior image model which incorporates anatomical information derived from registered MR images. A weighting parameter determines the relative emphasis given to the emission data and the prior model in the reconstruction. Details of this algorithm are presented as well as test reconstructions for real and simulated data. The performance of the algorithm was evaluated with respect to errors in prior anatomical information. The algorithm provided significant improvement in the quality of reconstructed images as compared with the ML-EM reconstruction technique. The reconstructed images had higher resolution as compared with the images obtained from MAP-like reconstructions which do not utilize anatomical information. The algorithm displayed robustness with respect to errors in prior anatomical information.
Braun, M 1996, 'Edge quality metric for arbitrary two-dimensional edges', Optical Engineering, vol. 35, no. 11, pp. 3222-3226.
A quality metric for an arbitrary 2D edge contour is presented. This metric is constructed from measured physical characteristics of the edge. It enables a comparison of performance of different edge detecting algorithms and provides a tool for the optimization of edge detectors. The metric was verified using the Canny's edge detector applied to a set of artificial images varying in image quality. © 1996 Society of Photo-Optical Instrumentation Engineers.
Narita, Y, Eberl, S, Iida, H, Hutton, BF, Braun, M, Nakamura, T & Bautovich, G 1996, 'Monte Carlo and experimental evaluation of accuracy and noise properties of two scatter correction methods for SPECT', Physics in Medicine and Biology, vol. 41, no. 11, pp. 2481-2496.View/Download from: Publisher's site
Scatter correction is a prerequisite for quantitative SPECT, but potentially increases noise. Monte Carlo simulations (EGS4) and physical phantom measurements were used to compare accuracy and noise properties of two scatter correction techniques: the triple-energy window (TEW), and the transmission dependent convolution subtraction (TDCS) techniques. Two scatter functions were investigated for TDCS: (i) the originally proposed mono-exponential function (TDCS(mono) and (ii) an exponential plus Gaussian scatter function (TDCS(Gauss)) demonstrated to be superior from our Monte Carlo simulations. Signal to noise ratio (S/N) and accuracy were investigated in cylindrical phantoms and a chest phantom. Results from each method were compared to the true primary counts (simulations), or known activity concentrations (phantom studies). 99mTc was used in all cases. The optimized TDCS(Gauss) method overall performed best, with an accuracy of better than 4% for all simulations and physical phantom studies. Maximum errors for TEW and TDCS(mono) of -30 and -22%, respectively, were observed in the heart chamber of the simulated chest phantom. TEW had the worst S/N ratio of the three techniques. The S/N ratios of the two TDCS methods were similar and only slightly lower than those of simulated true primary data. Thus, accurate quantitation can be obtained with TDCS(Gauss), with a relatively small reduction in S/N ratio.
Ardekani, BA, Braun, M, Hutton, BF, Kanno, I & Iida, H 1995, 'A fully automatic multimodality image registration algorithm', Journal of Computer Assisted Tomography, vol. 19, no. 4, pp. 615-623.View/Download from: Publisher's site
Objective: A fully automatic multimodality image registration algorithm is presented. The method is primarily designed for 3D registration of MR and PET images of the brain. However, it has also been successfully applied to CT-PET, MR-CT, and MR-SPECT registrations. Materials and Methods: The head contour is detected on the MR image using a gradient threshold method. The head region in the MR image is then segmented into a set of connected components using the K-means clustering algorithm. When the two image sets are registered, the segmentation of the MR image indirectly generates a segmentation of the PET image. The best registration is taken to be the one that optimizes the segmentation induced on the PET image. In this article, the K-means minimum variance criterion is used as a cost function, and the optimization is performed using the method of coordinate descent. Results: The algorithm was tested on 80 H2150 PET and MR image pairs from 10 subjects. Qualitatively correct results were obtained in all cases. With use of external markers visible in both image modalities, the average registration error was estimated to be <3 mm. Conclusion: The algorithm presented in this article requires no user interaction and can be applied to a wide range of registration problems. Quantitative and qualitative evaluations of the algorithm indicate a high degree of accuracy. © 1995 Lippincott-Raven Publishers, Philadelphia.
Ardekani, BA, Braun, M, Kanno, I & Hutton, BF 1994, 'Automatic detection of intradural spaces in mr images', Journal of Computer Assisted Tomography, vol. 18, no. 6, pp. 963-969.View/Download from: Publisher's site
Objective: An algorithm is presented for the automatic detection of intradural spaces in MR images of the human head. The primary motivation behind the present work has been to serve as a preprocessing step in automatic segmentation of brain tissue and CSF. A second objective was to use the algorithm in a fully automatic PET-MR registration algorithm. Materials and Methods: The method is primarily designed for, and requires, dual echo (Tl- and T2-weighted) MR images with transaxial orientations. The algorithm consists of three main stages. First, the head contour is detected using a series of low-level image-processing techniques. In the second stage, the pixels inside the head contour are clustered into a number of classes using the K-means algorithm. Finally, the extradural connected components are eliminated based on a number of heuristics. Results: Test results are presented for 10 MR image sets consisting of 197 slices. As a quantitative measure of accuracy, manual segmentations were performed by radiologists on a number of slices and compared with the results obtained automatically. Conclusion: Visual inspection and quantitative validation of the results indicate that the algorithm accurately detects the intradural spaces in MR images. This is an important step in fully automatic segmentation and registration of MR images. © 1994 Raven Press, Ltd., New York.
Fulton, RR, Hutton, BF, Braun, M, Ardekani, B & Larkin, R 1994, 'Use of 3D reconstruction to correct for patient motion in SPECT', Physics in Medicine and Biology, vol. 39, no. 3, pp. 563-574.View/Download from: Publisher's site
Patient motion occurring during data acquisition in single photon emission computed tomography (SPECT) can cause serious reconstruction artefacts. The authors have developed a new approach to correct for head motion in brain SPECT. Prior to motion, projections are assigned to conventional projections. When head motion occurs, it is measured by a motion monitoring system, and subsequent projection data are mapped to 'virtual' projections. The appropriate position of each virtual projection is determined by applying the converse of the patient's accumulated motion to the actual camera projection. Conventional and virtual projections, taken together, form a consistent set that can be reconstructed using a three-dimensional (3D) algorithm. The technique has been tested on a range of simulated rotational movements, both within and out of the transaxial plane. For all simulated movements, the motion corrected images exhibited better agreement with a motion free reconstruction than did the uncorrected images. This technique may help to overcome one of the major remaining limitations on image quality and quantitative accuracy in SPECT.
Ching, NH, Rosenfeld, D & Braun, M 1992, 'Two-Dimensional Phase Unwrapping Using a Minimum Spanning Tree Algorithm', IEEE Transactions on Image Processing, vol. 1, no. 3, pp. 355-365.View/Download from: Publisher's site
Phase unwrapping refers to the determination of phase from modulo 2π data. Some of the phase data may not be reliable (e.g., where the magnitude approaches zero or where the signal-to-noise ratio is poor). In two dimensions, this is equivalent to confining the support of the phase function to one or more arbitrarily shaped regions. A phase unwrapping algorithm is presented which works for two-dimensional (2-D) data known only within a set of nonconnected regions with possibly nonconvex boundaries. The algorithm includes the following steps: segmentation to identify connectivity, phase unwrapping within each segment using a Taylor series expansion, phase unwrapping between disconnected segments along an optimum path, and filling of phase information voids. The optimum path for intersegment unwrapping is determined by a minimum spanning tree algorithm. Although the algorithm is applicable to any 2-D data, the main application addressed is magnetic resonance imaging (MRI) where phase maps are useful in determining the distributions of the applied magnetic field, inherent chemical shifts of the object, and the object's magnetic susceptibility. © 1992 IEEE
Wang, H, Rosenfeld, D, Braun, M & Yan, H 1992, 'Compression and reconstruction of MRI images using 2D DCT', Magnetic Resonance Imaging, vol. 10, no. 3, pp. 427-432.View/Download from: Publisher's site
In magnetic resonance imaging (MRI), the original data are sampled in the spatial frequency domain. The sampled data thus constitute a set of discrete Fourier transform (DFT) coefficients. The image is usually reconstructed by taking inverse DFT. The image data may then be compressed using the discrete cosine transform (DCT). We present here a method of treating the data that combines two procedures, image reconstruction and data compression. This method may be particularly useful in medical picture archiving and communication systems (PACS) where both image reconstruction and compression are important issues. © 1992.
Wang, H, Rosenfeld, D, Braun, M & Yan, H 1992, 'Efficient algorithm for MR image reconstruction and compression', Australasian Physical and Engineering Sciences in Medicine, vol. 15, no. 3, pp. 133-137.
In magnetic resonance imaging (MRI), the original data are sampled in the spatial frequency domain. The sampled data thus constitute a set of discrete Fourier transform (DFT) coefficients. The image is usually reconstructed by taking inverse DFT. The image data may then be efficiently compressed using the discrete cosine transform (DCT). We present here a method of using DCT to treat the sampled data, which combines two procedures, image reconstruction and data compression. This method may be particularly useful in medical picture archiving and communication systems (PACS) where both image reconstruction and compression are important issues.
Yan, H & Braun, M 1991, 'Image reconstruction from Fourier domain data sampled along a zig‐zag trajectory', Magnetic Resonance in Medicine, vol. 18, no. 2, pp. 405-410.View/Download from: Publisher's site
When the conventional Fourier transform (FT) algorithm is applied to reconstruct a magnetic resonance (MR) image from data sampled along a zig‐zag trajectory in the Fourier space, the nonuniform sampling in the spatial frequency direction may give rise to artifacts. In this paper the nature of the artifacts is analyzed and an alternative reconstruction algorithm is developed to produce artifact‐free images. Methods for reducing noise level in the reconstructed image are discussed. Our approach is compared with another method based on the interlace sampling theorem. © 1991 Academic Press. Inc. Copyright © 1991 Wiley‐Liss, Inc., A Wiley Company
Wang, H, Rosenfeld, D & Braun, M 1989, 'Pulse, a PC-based graphics pulse sequence editor for NMR and MRI', Australasian Physical and Engineering Sciences in Medicine, vol. 12, no. 2, pp. 69-72.
A flexible, personal computer (PC) based, screen-graphics oriented pulse sequence editor called PULSE has been developed for nuclear magnetic resonance (NMR) spectroscopy and magnetic resonance imaging (MRI). PULSE is used to set such NMR spectroscopic parameters as the delay and duration of rf transmit and receive gates, rf phase, sampling times, and such imaging parameters as rf pulse shape and gradient waveforms. The output of PULSE is a set of programs that can be loaded into a hardware pulse programmer. With PULSE, any desired NMR or MRI pulse sequence can be programmed quickly and easily.
Blum, MJ, Braun, M & Rosenfeld, D 1987, 'Fast magnetic resonance imaging using spiral trajectories', Proceedings of SPIE - The International Society for Optical Engineering, vol. 767, pp. 40-46.View/Download from: Publisher's site
Acquisition times in magnetic resonance imaging (MRI) are typically in the order of minutes. For an image of 256 × 256 pixels, the standard Fourier reconstruction technique used in most commercial imaging systems requires 256 separate free induction decay (FID) signals. While the FID signal itself is of relatively short duration, the successive FID signals are separated by long delays, of the order of seconds, to permit substantial relaxation of the signal before the next excitation. The resultant long acquisition times give rise to motion artefacts, preclude dynamic imaging and keep the patient throughput low. In this paper, we investigate a fast imaging scheme which uses spiral trajectories in the spatial frequency domain. The entire domain can be sampled in a short time, requiring as few as one FID acquisition. The scheme requires time varying gradients having the form of a ramped sinusoid. Several reconstruction methods are considered for forming the image from the spatial frequency domain data. The possibility of using multiple spirals to deal with the rapid decay of the FID signal is also examined in detail. © 1987 SPIE.
Blum, MJ, Braun, M & Rosenfeld, D 1987, 'FAST MAGNETIC RESONANCE IMAGING USING SPIRAL TRAJECTORIES.', Australasian Physical and Engineering Sciences in Medicine, vol. 10, no. 2, pp. 79-87.
Acquisition times in magnetic resonance imaging are typically in the order of minutes. These long acquisition times give rise to motion artefacts, preclude dynamic imaging and keep the patient throughput low. The authors investigate a fast imaging scheme which uses spiral trajectories in the spatial frequency domain. The entire domain can be sampled in a short time, requiring as few as one free induction decay (FID) acquisition. The scheme requires time varying gradients having the form of a ramped sinusoid. Several reconstruction methods are considered for forming the image from the spatial frequency domain data. The possibility of using multiple spirals to deal with the rapid decay of the FID signal is also examined in detail.
Braun, M & Rosenfeld, D 1987, 'COMPUTER SYSTEM ARCHITECTURE FOR NMR IMAGING.', Australasian Physical and Engineering Sciences in Medicine, vol. 10, no. 2, pp. 64-71.
NMR imaging makes extensive use of computers in data acquisition, image reconstruction, processing and display. Furthermore, in order to fully utilise the capability of the method, flexible control must be exercised over the radiofrequency and gradient pulse sequence. The requirements of operational flexibility and speed favour a distributed architecture with dedicated programmable devices taking on separate tasks under the overall control of a host computer. A fast, versatile pulse programmer is required to provide real-time control of a number of rapidly varying, experimental parameters. A state-memory-based, pulse programmer, having the architecture of a computer control unit, affords the optimum design for imaging applications. In image processing, array processors have a considerable speed advantage and will find increasing use in clinical imaging systems.
Ching, NH, Rosenfeld, D & Braun, M 1987, 'NOVEL COMPENSATING FILTER FOR RESOLUTION ENHANCEMENT IN ULTRASONIC IMAGING.', Australasian Physical and Engineering Sciences in Medicine, vol. 10, no. 2, pp. 72-78.
This study investigates the compensation of the spreading of ultrasonic pulses due to frequency-dependent attenuation by tissues. The effect of pulse spreading progressively degrades the axial resolution with increasing depth. A novel technique, using a time varying filter made up of a bank of tapped delay lines is proposed to compensate for this effect. The performance of the filter is evaluated by means of computer simulation and the limits of its application are discussed. Apart from improving the axial resolution (and, hence, the accuracy of axial measurements), the compensating filter could be used in implementing tissue characterisation algorithms based on attenuation data.
Ching, NH, Rosenfeld, D & Braun, M 1987, 'Resolution Enhancement in Ultrasonic Imaging by a Time-Varying Filter', Proceedings of SPIE - The International Society for Optical Engineering, vol. 768, pp. 79-84.View/Download from: Publisher's site
The study reported here investigates the use of a time-varying filter to compensate for the spreading of ultrasonic pulses due to the frequency dependence of attenuation by tissues. The effect of this pulse spreading is to degrade progressively the axial resolution with increasing depth. The form of compensation required to correct for this effect is impossible to realize exactly. A novel time-varying filter utilizing a bank of bandpass filters is proposed as a realizable approximation of the required compensation. The performance of this filter is evaluated by means of a computer simulation. The limits of its application are discussed. Apart from improving the axial resolution, and hence the accuracy of axial measurements, the compensating filter could be used in implementing tissue characterization algorithms based on attenuation data. © 1987 SPIE.
Town, G, Ching, N, Braun, M & Rosenfeld, D 1987, 'Design and Construction of "SUMRIS" -The Sydney University Magnetic Resonance Imaging System', Proceedings of SPIE - The International Society for Optical Engineering, vol. 767, pp. 55-60.View/Download from: Publisher's site
A magnetic resonance imaging system for investigation and development of nuclear magnetic resonance (NMR) imaging techniques is being constructed by the NMR Imaging Group within the School of Electrical Engineering at the University of Sydney. The system will also be capable of spatially-localized in-vivo spectroscopy. Primary requirements of a research instrument are flexibility and low cost, precluding purchase of a commercial system and necessitating in-house construction. The appropriate design strategy adopted was to start by building a simple NMR apparatus and expand it in steps, progressing towards a complete imaging and spectroscopy system. © 1987, SPIE.
Town, G, Ching, N, Braun, M & Rosenfeld, D 1987, 'DESIGN AND CONSTRUCTION OF 'SUMRIS' - THE SYDNEY UNIVERSITY MAGNETIC RESONANCE IMAGING SYSTEM.', Australasian Physical and Engineering Sciences in Medicine, vol. 10, no. 2, pp. 58-63.
Magnetic resonance imaging (MRI) is a highly versatile medical imaging modality which uses nuclear magnetic resonance (NMR) techniques to obtain the spatial distribution of a combination of tissue-dependent NMR parameters. The NMR Imaging Group at the School of Electrical Engineering, University of Sydney, has carried out a number of investigations into various aspects of magnetic resonance imaging, primarily in signal processing and image formation algorithms. To enable further development of these and other techniques it was found necessary to construct SUMRIS - the Sydney University Magnetic Resonance Imaging System, a flexible imaging and spectroscopy experimental facility. The appropriate design strategy adopted was to start by building a simple NMR apparatus and expand it in steps, progressing towards a complete imaging and spectroscopy system.
NMR images of distinct sections through a three‐dimensional phantom were obtained using steady magnetic fields much less uniform than those generally considered necessary for NMR imaging, but whose substantial gradient served to define the slices to be imaged. A sensitive point technique and a modified spin‐echo pulse sequence were used. NMR images are generally produced in highly uniform steady magnetic fields, to which switched gradients are added to achieve spatial selection. In this note we show that NMR images can be produced in a substantially non‐uniform field, provided the field is shaped so that the gradient of the steady field can be used to define the slice to be imaged. Copyright © 1985, Wiley Blackwell. All rights reserved
Comparisons of image quality were made for several rare-earth film-screen combinations under conditions similar to those used in chest radiography. The characteristic curve, modulation transfer function, and Wiener spectrum were measured. Observer response studies were carried out to compare the limits of perceived resolution. Signal-to-noise considerations based on the measured characteristics favor a fast film-slow screen pair over a slow film-fast screen pair, consistent with the observer study results. The interrelationship of the physical parameters in determining image quality and the limitations of the method are discussed.
Hutton, B, Braun, M & Slomka, P 2006, 'Image registration techniques in nuclear medicine imaging' in Zaidi, H (ed), Quantitative Analysis in Nuclear Medicine Imaging, Springer, New York, USA, pp. 272-307.View/Download from: Publisher's site
Neclear medicine has a long tradition of incorporating quantitative analysis in its diagnostic procedures. Until recently, the analysis was based on radionuclide images as the sole input although the importance of the complementary information available from other modalities or from earlier scans has long been recognised. Indeed, qualitative correlation between images, based on anatomical expertise, has always been part of the repertoire of nuclear medicine clinician. However, spatial distortiom netween images, caused largely by differences in posture and acquisition technique, prevented the extension of these techniques to quantitative analysis. Recent advances in image registration software and hardware have made it increasingly possible to utilise that complementary image information in a clinical setting.
Braun, M, Schulte, J & Davila, YC 2015, 'Reciprocal Peer Teaching for Problem-Solving Teams in a Senior-Year Science Course', Proceedings of the Australian Conference for Science and Mathematics Education, Australian Conference for Science and Mathematics Education 2015, The University of Sydney, Institute for Innovation in Science & Mathematics Education, Perth, Australia, pp. 13-14.
Kirkup, L, Varadharajan, M, Braun, M, Buffler, A & Lubben, F 2014, 'Matching the background of demonstrators with those of their students: does it make a difference?', Students Transitions Achievement Retention and Success, Melbourne.
Eberl, S & Braun, M 1999, 'Intra- and inter-modality registration of functional and anatomical clinical images', Proceedings of SPIE - The International Society for Optical Engineering, pp. 102-114.
Image registration techniques spatially register clinical images of patients performed either at different times with the same modality (intra-modality) or with different modality (inter-modality), to facilitate assessment of change and to take full advantage of the frequently complementary information provided by the imaging modalities. Inter-subject registration permits comparison to normal data bases and averaging of data from several subjects to improve statistical significance. Image registration is well established in the brain since the skull limits deformation of the brain between studies and the use of rigid body transformation can usually be justified for intra-subject registrations. A large range of image registration algorithms, ranging from completely manual to fully automatic, have been developed. These can be classified into external methods, which typically use fiducial markers, intrinsic techniques, which rely on the information contained in the patient image data and non-image based methods, which use information external to the data being registered. The technique of choice depends on the specific requirements of the application and it is unlikely that a single 'best' technique can meet sometimes conflicting requirements (e.g. accuracy, speed, ease of use etc). While considerable progress has been made in image registration outside the brain, considerable challenges still remain. In this paper, we present the basic principles of image registration and practical issues arising from our experience with routine clinical use of image registration over several years.
Abbott, P & Braun, M 1997, 'Segmentation of ultrasound image data by two dimensional autoregressive modelling', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 672-679.View/Download from: Publisher's site
© Springer-Verlag Berlin Heidelberg 1997. In this paper we treat ultrasound image data as a two dimensional autoregressive (AR) signal. The image is modelled as consisting of distinct regions each described by one of a small number of AR models. Segmentation is performed by maximising the image likelihood function, which takes on a convenient form due to the AR model. Image data is presented to the algorithm in complex amplitude form. Results from application of this method to a cardiac phantom data set are presented.
Som, S, Hutton, BF & Braun, M 1997, 'Properties of minimum cross-entropy reconstruction of emission tomography with a morphologically based prior', IEEE Nuclear Science Symposium & Medical Imaging Conference, pp. 1738-1742.
We have studied the properties of a minimum cross-entropy (MXE) algorithm for emission tomography reconstruction with a morphological prior. MXE is formulated with two terms: a maximum likelihood expectation maximization term and a penalty term for regularization within morphologically defined boundaries. The relative emphasis put on the two competing terms is controlled by the regularization constant β. Edge resolution and noise were compared for reconstructions with and without corresponding prior. The prior leads to significant edge enhancement with edge resolution converging to a theoretical limit independent of β. Normalized standard deviation (NSD) and resolution both illustrate that regularization within boundaries behaves predictably with more smoothing for larger β. Application of ordered subsets (OS) was also investigated. For OS, edge enhancement is fully preserved but NSD increases for low subset size. Results demonstrate that OS is applicable to MXE provided subset size is greater than 4. OS-MXE has appealing properties for regularized reconstruction.
Zhang, Z & Braun, M 1997, 'Fully 3D active surface models with self-inflation and self-deflation forces', Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 85-90.
In this paper, we propose fully 3D active surface models for image segmentation. Our models are capable of fitting a diverse range of region shapes. They have low sensitivity to initial shape and position. We design self-inflation/deflation forces, which cooperate naturally with gradient forces. They permit the active surface to travel a long distance without the aid of any external forces. They are easily controlled in both their direction and magnitude. The models produce accurate segmentation when tested with synthetic and real images. They manifest robustness to image noise and imperfect image data. Importantly, they are capable of converging to the correct boundary even if the initial estimate is not close.
Zhang, Z, Braun, M & Abbott, P 1997, 'A new deformable model for 3D image segmentation', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), pp. 239-246.View/Download from: Publisher's site
© Springer-Verlag Berlin Heidelberg 1997. A fully 3D active surface model is presented with self-inflation and self-deflation forces. The model makes full use of 3D image information, deforms locally and allowes strong deformation. The self-inflation and self-deflation forces enable the active surface to travel a long distance without the help from any external forces. We introduce a method of adapting model parameters, which enables our model to bypass some noise and irrelevant edge points. The model is tested with synthetic and real images. Accurate segmentation results are obtained in the presence of image noise and imperfect image data. Importantly, the model is capable of converging to the correct boundary even if the initial estimate is not close. Computational efficiency of segmentation with our model is addressed.
Ardekani, BA, Braun, M & Hutton, BF 1993, 'Improved quantification with the use of anatomical information in PET image reconstruction', Annals of Nuclear Medicine.
Robb, RA, Ellisman, MH, Bosan, S, Harris, TR, Poon, CS, Braun, M, Ginige, A, Tusda, R, Ueno, S, Truyen, B, Cornelis, J & Vandervelder, P 1993, 'Proceedings of the 15th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Part 3 (of 3)', Proceedings of the Annual Conference on Engineering in Medicine and Biology.
Recent development of advanced quantitative methods to fully analyze the intrinsic information contained in images, have begun to unearth the rich treasures they contain and to exploit their full scientific, educational and/or biomedical value. structural alterations found in the brains of patients with Alzheimer's disease. The second project concerns the three-dimensional reconstruction of the internal membrane system of neurons involved in the regulation of intracellular calcium. The final project involves the use of tomographic reconstruction techniques to visualize components of the nervous systems in thick sections imaged with the aid of intermediate high voltage electron microscopy. parameter sensitivity directions to determine identifiability for a capillary exchange model. The method is based on a vector definition of model sensitivity to parameter values. The results are images which provide immediate recognition of the identifiability of the parameters over wide ranges. processing and analysis. We propose a smoothing algorithm which preserves edges and thus fine structures common in biomedical images. The algorithm is based on averaging over an independently generated mask at each pixel. The mask generation takes into account local image information and conforms to tissue boundaries. Comparison of the proposed algorithm with other common smoothing techniques favors our algorithm. with the psychological properties in central vision, however, apparent movement in peripheral vision have not been clearly understood. This study focuses on the properties of the perception of apparent movement between central and peripheral vision. improve upon existing image reconstruction techniques in Electrical Impedance Tomography. The dynamics of the network are surprisingly simple and retain the possibility of analog implementation. Compared to the algorithms currently used in EIT, the proposed method differs in its emphasis on relaxation processing and rigorous mathema...