
Overview
Background
Shakes an imaging expert that leads a strong deep learning, artificial intelligence (AI) focused research team interested in medical image analysis and signal/image processing applied to many areas of science and medicine. He received his Ph.D in Theoretical Physics from Monash University, Melbourne and has been involved in applying machine learning in medical imaging for over a decade.
Shakes’ past work has involved developing shape model-based algorithms for knee, hip and shoulder joint segmentation that is being developed and deployed as a product on the Siemens syngo.via platform. More recent work involves deep learning based algorithms for semantic segmentation and manifold learning of imaging data. Broadly, he is interested in understanding and developing the mathematical basis of imaging, image analysis algorithms and physical systems. He has developed algorithms that utilise exotic mathematical structures such as fractals, turbulence, group theoretic concepts and number theory in the image processing approaches that he has developed.
He is currently a Senior Lecturer and leads a team of 20+ researchers working image analysis and AI research across healthcare and medicine. He currently teaches the computer science courses Theory of Computation and Pattern Recognition and Analysis.
Availability
- Dr Shakes Chandra is:
- Available for supervision
Fields of research
Qualifications
- Doctor of Philosophy, Monash University
Research interests
-
Magnetic Resonance Imaging
Making MRI faster and more affordable through better image reconstruction, processing and analysis.
-
Image Processing
Image reconstruction, segmentation and registration.
-
Deep learning
Dimensionality reduction, machine learning and Artificial Intelligence
-
Fractals and Chaos
Applying fractals and chaos to image processing and computer science.
-
Number Theory
Applying number theory to image processing and computer science.
-
Medical Image Analysis
Medical image segmentation and shape analysis
Works
Search Professor Shakes Chandra’s works on UQ eSpace
2013
Journal Article
Automated bone segmentation from large field of view 3D MR images of the hip joint
Xia, Ying, Fripp, Jurgen, Chandra, Shekhar S., Schwarz, Raphael, Engstrom, Craig and Crozier, Stuart (2013). Automated bone segmentation from large field of view 3D MR images of the hip joint. Physics in Medicine and Biology, 58 (20), 7375-7390. doi: 10.1088/0031-9155/58/20/7375
2013
Conference Publication
Direct Inversion of Mojette Projections
Svalbe, Imants, Kingston, Andrew, Guedon, Jeanpierre, Normand, Nicolas and Chandra, Shekhar S. (2013). Direct Inversion of Mojette Projections. 20th IEEE International Conference on Image Processing, ICIP 2013, Melbourne , Australia, 15 - 18 September 2013. Piscataway, NJ United States: IEEE. doi: 10.1109/ICIP.2013.6738214
2012
Journal Article
Recovering missing slices of the discrete fourier transform using ghosts
Chandra, Shekhar S., Svalbe, Imants D., Guedon, Jeanpierre, Kingston, Andrew M. and Normand, Nicolas (2012). Recovering missing slices of the discrete fourier transform using ghosts. IEEE Transactions on Image Processing, 21 (10) 6226457, 4431-4441. doi: 10.1109/TIP.2012.2206033
2012
Journal Article
Patient specific prostate segmentation in 3-D magnetic resonance images
Chandra, Shekhar S., Dowling, Jason A., Shen, Kai-Kai, Raniga, Parnesh, Pluim, Josien P. W., Greer, Peter B., Salvado, Olivier and Fripp, Jurgen (2012). Patient specific prostate segmentation in 3-D magnetic resonance images. IEEE Transactions On Medical Imaging, 31 (10) 6257497, 1955-1964. doi: 10.1109/TMI.2012.2211377
2012
Conference Publication
Unilateral hip joint segmentation with shape priors learned from missing data
Chandra, Shekhar, Xia, Yinq, Engstrom, Craig, Schwarz, Raphael, Lauer, Lars, Crozier, Stuart, Salvado, Olivier and Fripp, Jurgen (2012). Unilateral hip joint segmentation with shape priors learned from missing data. 9th IEEE International Symposium on Biomedical Imaging (ISBI), Barcelona, Spain, 2-5 March 2012. Piscataway, NJ, United States: IEEE. doi: 10.1109/ISBI.2012.6235909
2012
Conference Publication
Morphology-based interslice interpolation on manual segmentations of joint bones and muscles in MRI
Yang, Zhengyi, Crozier, Stuart, Engstrom, Craig, Xia, Ying, Neubert, Ales, Brancato, Tania, Schwarz, Raphael, Lauer, Lars, Fripp, Jurgen, Chandra, Shekhar and Salvado, Olivier (2012). Morphology-based interslice interpolation on manual segmentations of joint bones and muscles in MRI. 2012 International Conference on Digital Image Computing Techniques and Applications (DICTA), Fremantle, WA, Australia, 3-5 December 2012. Piscataway, NJ, United States: IEEE. doi: 10.1109/DICTA.2012.6411678
2012
Conference Publication
Automated bone segmentation and bone-cartilage interface extraction from MR images of the hip
Xia, Ying, Chandra, Shakes, Salvado, Oliver, Fripp, Jurgen, Schwartz, Raphael, Lauer, Lars, Engstrom, Craig M. and Crozier, Stuart (2012). Automated bone segmentation and bone-cartilage interface extraction from MR images of the hip. International Society for Magnetic Resonance in Medicine, Melbourme, VIC, Australia, 5-11 May 2012.
2011
Conference Publication
Automated MR hip bone segmentation
Xia, Ying, Chandra, Shakes, Salvado, Olivier, Fripp, Jurgen, Schwarz, Raphael, Lauer, Lars, Engstrom, Craig and Crozier, Stuart (2011). Automated MR hip bone segmentation. International Conference on Digital Image Computing Techniques and Applications (DICTA), Noosa, QLD, Australia, 6-8 December 2011. Piscataway, NJ, United States: IEEE. doi: 10.1109/DICTA.2011.13
2011
Conference Publication
Growth of Discrete Projection Ghosts Created by Iteration
Svalbe, Imants and Chandra, Shekhar (2011). Growth of Discrete Projection Ghosts Created by Iteration. 16th International Conference on Discrete Geometry for Computer Imagery, Nancy France, 6 - 8 April 2011. Heidelberg, Germany: Springer. doi: 10.1007/978-3-642-19867-0_34
2011
Conference Publication
Fast automatic multi-atlas segmentation of the prostate from 3D MR images
Dowling, Jason A., Fripp, Jurgen, Chandra, Shekhar, Pluim, Josien P. W., Lambert, Jonathan, Parker, Joel, Denham, James, Greer, Peter B. and Salvado, Olivier (2011). Fast automatic multi-atlas segmentation of the prostate from 3D MR images. 14th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2011), Toronto, Canada, 18-22 September 2011. Heidelberg, Germany: Springer. doi: 10.1007/978-3-642-23944-1_2
2011
Conference Publication
Automatic segmentation of the prostate in 3D magnetic resonance images using case specific deformable models
Chandra, Shekhar, Dowling, Jason, Shen, Kaikai, Pluim, Josien, Greer, Peter, Salvado, Olivier and Fripp, Jurgen (2011). Automatic segmentation of the prostate in 3D magnetic resonance images using case specific deformable models. International Conference on Digital Image Computing: Techniques and Applications, DICTA 2011, Noosa Heads, Qld., Australia, 6-8 December 2011. Piscataway, NJ, United States: I E E E. doi: 10.1109/DICTA.2011.10
2010
Conference Publication
On constructing minimal ghosts
Svalbe, Imants, Nazareth, Nikesh, Chandra, Shekhar and Normand, Nicolas (2010). On constructing minimal ghosts. International Conference on Digital Image Computing: Techniques and Applications, DICTA 2010, Sydney, NSW Australia, 01 - 03 December 2010. Piscataway, NJ United States: I E E E. doi: 10.1109/DICTA.2010.56
2010
Other Outputs
Circulant theory of the Radon transform
Chandra, Shekhar Suresh (2010). Circulant theory of the Radon transform. PhD Thesis, Faculty of Science, School of Physics, Monash University.
2009
Conference Publication
A fast number theoretic finite radon transform
Chandra, S. and Svalbe, I. (2009). A fast number theoretic finite radon transform. Digital Image Computing: Techniques and Applications, DICTA 2009, Melbourne, VIC Australia, 1 - 3 December 2009. Piscataway, NJ United States: I E E E. doi: 10.1109/DICTA.2009.67
2008
Conference Publication
An exact, non-iterative Mojette inversion technique utilising ghosts
Chandra, Shekhar, Svalbe, Imants and Guedon, Jean-Pierre (2008). An exact, non-iterative Mojette inversion technique utilising ghosts. 14th International Conference on Discrete Geometry for Computer Imagery, Lyon, France, 16 - 18 April 2008. Heidelberg, Germany: Springer. doi: 10.1007/978-3-540-79126-3_36
2008
Conference Publication
A method for removing cyclic artefacts in discrete tomography using latin squares
Chandra, Shekhar and Svalbe, Imants (2008). A method for removing cyclic artefacts in discrete tomography using latin squares. 19th International Conference on Pattern Recognition (ICPR 2008), Tampa, Fl United States, 8 - 11 Dec 2008. Washington, DC United States: I E E E Computer Society. doi: 10.1109/ICPR.2008.4761615
2006
Journal Article
Quantised angular momentum vectors and projection angle distributions for discrete radon transformations
Svalbe, Imants, Chandra, Shekhar, Kingston, Andrew and Guedon, Jean-Pierre (2006). Quantised angular momentum vectors and projection angle distributions for discrete radon transformations. Discrete Geometry for Computer Imagery, Proceedings, 4245, 134-145.
2006
Conference Publication
Quantised angular momentum vectors and projection angle distributions for discrete radon transformations
Svalbe, Imants, Chandra, Shekhar, Kingston, Andrew and Guédon, Jean-Pierre (2006). Quantised angular momentum vectors and projection angle distributions for discrete radon transformations. 13th International Conference on Discrete Geometry for Computer Imagery, DGCI 2006, , , October 25, 2006-October 27, 2006. Springer Verlag. doi: 10.1007/11907350_12
Funding
Current funding
Past funding
Supervision
Availability
- Dr Shakes Chandra is:
- Available for supervision
Before you email them, read our advice on how to contact a supervisor.
Available projects
-
Machine learning applied to 3D magnetic resonance images
Magnetic resonance (MR) imaging has become an important non-invasive radiological modality for various clinical applications, such as cartilage assessment for Osteoarthritis and treatment planning for prostate cancer. MR images in 3D, while providing a wealth of anatomical information, including bones and soft tissue, are difficult to analyse due to the presence of a large number of complex structures. Thus, extracting meaningful clinical information without human interaction is a challenging task. Developing such automatic methods are important in order to reduce human errors and the time taken by clinicians in completing mundane tasks, such as marking or delineating 3D images by hand, from hours to just a few minutes by utilising computers.
In this project, the student will develop novel algorithms to solve segmentation and detection problems for MR imaging that could possibly be deployed to MRI scanners and may eventually used for diagnostic purposes. The project will involve applying computer vision and machine learning techniques (including deep learning) to MR image processing and analysis.
Supervision history
Current supervision
-
Doctor Philosophy
Magnetic Resonance Image Processing with Artificial Intelligence
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Dr Hongfu Sun
-
Doctor Philosophy
Manifold Learning for Magnetic Resonance Imaging
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
Doctor Philosophy
Manifold Learning for Magnetic Resonance Imaging
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
Doctor Philosophy
Foundational models for visual recognition
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
Doctor Philosophy
Deep Learning Applied to Medical Image Analysis
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
Doctor Philosophy
Deep Learning Strategies for Enhanced mTBI Diagnosis Using Clinical and CT Data
Principal Advisor
-
Doctor Philosophy
Towards Analysis of Contextual Melanoma Indicators and Identification of Total-Body Ugly Duckling Lesions with Deep Neural Networks
Principal Advisor
Other advisors: Associate Professor Mahsa Baktashmotlagh
-
Doctor Philosophy
Deployable and Explainable Deep Learning Architectures: Navigating Clinical Challenges in Medical Image Segmentation
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
Doctor Philosophy
Automated Detection and Classification of Suspicious Naevi in Dermoscopy Images Through Artificial Intelligence
Principal Advisor
Other advisors: Professor Peter Soyer, Professor Monika Janda
-
Doctor Philosophy
Medical Shape Analysis of 3D MRI Segmentation with Deep Learning
Principal Advisor
Other advisors: Emeritus Professor Stuart Crozier
-
Doctor Philosophy
Improving Feature Extraction and Feature Interpretability for Deep Learning in Medical Image Analysis
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
Doctor Philosophy
Novel deep learning approaches to understanding human diseases
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
Doctor Philosophy
Advanced Deep Learning Approaches for Improving Diagnosis and Prognosis in Brain Disease
Associate Advisor
Other advisors: Associate Professor Fatima Nasrallah
-
Doctor Philosophy
Using 3D total body imaging to study the spatial distribution of naevi and melanoma
Associate Advisor
Other advisors: Professor Peter Soyer, Professor Monika Janda
-
Doctor Philosophy
Advanced Deep Learning Approaches for Improving Diagnosis and Prognosis in Brain Disease
Associate Advisor
Other advisors: Associate Professor Fatima Nasrallah
-
Doctor Philosophy
New deep learning based methods for assessment of sleep apnea severity and risk of related short and long term health consequences
Associate Advisor
Other advisors: Professor Juha Toyras, Associate Professor Philip Terrill
-
Doctor Philosophy
Artifical intelligence approaches for diagnosing and phenotyping sleep disorders
Associate Advisor
Other advisors: Professor Juha Toyras, Associate Professor Philip Terrill
-
Doctor Philosophy
Artifical intelligence approaches for diagnosing and phenotyping sleep disorders
Associate Advisor
Other advisors: Professor Juha Toyras, Associate Professor Philip Terrill
-
Doctor Philosophy
Virtual Agricultural Imaging and Sensing through Artificial Intelligence and Computer Vision
Associate Advisor
Other advisors: Professor Scott Chapman
-
Doctor Philosophy
Establishment of a National Anterior Cruciate Ligament (ACL) Registry in Australia
Associate Advisor
Other advisors: Associate Professor Craig Engstrom
-
Doctor Philosophy
New deep learning based methods for assessment of sleep apnea severity and risk of related short and long term health consequences
Associate Advisor
Other advisors: Professor Juha Toyras, Associate Professor Philip Terrill
-
Doctor Philosophy
Development of a model for prognostication of patient outcome following traumatic brain injury
Associate Advisor
Other advisors: Associate Professor Fatima Nasrallah
-
Doctor Philosophy
Machine learning methods for visualisation and quantification of nephrons with MRI.
Associate Advisor
Other advisors: Dr Nyoman Kurniawan
Completed supervision
-
2024
Doctor Philosophy
Deployable and Interpretable Deep Learning Architectures: Navigating Clinical Challenges in Medical Image Segmentation
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
2024
Doctor Philosophy
Efficient Image Representations for Compressed Sensing MRI
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Professor Feng Liu, Professor Markus Barth
-
2024
Doctor Philosophy
Deep Learning Strategies for Enhanced mTBI Diagnosis using Clinical and CT Data
Principal Advisor
-
2024
Doctor Philosophy
Towards Analysis of Contextual Melanoma Indicators and Identification of Total-Body Ugly Duckling Lesions with Deep Neural Networks
Principal Advisor
Other advisors: Associate Professor Mahsa Baktashmotlagh
-
2023
Doctor Philosophy
Towards Efficient Graph Neural Networks for Optimizing Illicit Dark Web Interventions
Principal Advisor
Other advisors: Professor Marius Portmann
-
2022
Doctor Philosophy
Automated Assessment of Cartilage Composition and Cam Morphology using Magnetic Resonance Images of the Hip Joint
Principal Advisor
Other advisors: Associate Professor Craig Engstrom, Emeritus Professor Stuart Crozier
-
2020
Doctor Philosophy
Computer aided lesion detection, segmentation and characterization on mammography and breast MRI
Principal Advisor
Other advisors: Emeritus Professor Stuart Crozier
-
2023
Doctor Philosophy
In vitro evaluation of porous PHBV-based scaffolds for tissue regeneration application
Associate Advisor
Other advisors: Dr Mingyuan Lu
-
2022
Doctor Philosophy
Improving deep learning-based fast MRI with pre-acquired image guidance
Associate Advisor
Other advisors: Emeritus Professor Stuart Crozier, Professor Feng Liu
Media
Enquiries
For media enquiries about Dr Shakes Chandra's areas of expertise, story ideas and help finding experts, contact our Media team: