Associate Professor Viktor Vegh's interests lie in the research and development of technologies to improve the diagnosis, intervention, and prevention of health conditions. Essentially, he creates technologies to make measurements, or use already collected measurements to make inferences. He aims to produce quantitative information using technologies founded on mathematical methods or data driven approaches, including machine learning and deep learning. The primary technologies of interest for him have been those linked with human medical imaging (primarily, MRI, PET, SPECT, CT). More recently, he has extended his effort into other types of data, including medical records, and genetics, amongst others.
Associate Professor Vegh is a trained mathematician with an interest in neuroimaging and more broadly imaging of the human body. He has developed MRI instrumentation (gradient coils, superconducting and permanent magnets, radio frequency coils). For more than a decade, he has been developing low field MRI instrumentation with the aim of making MRI technology more accessible, portable, all at a lower cost. In parallel he has designed and implemented medical imaging protocols to improve clinical imaging workflow and the information which can be extracted from images. Topics of interest include tissue microstructure imaging using MRI, non-traditional relaxation and diffusion processes in MRI, and data driven methods for information mapping. Techniques employed within the team have now been extended from mathematical foundations to machine learning and deep learning. The Vegh Group is increasingly focusing on how our methods can be applied for prevention of disease, not just at the time of intervention.
Industry
The Vegh Group is currently working with Intellidesign Pty Ltd (design and manufacturing company in Brisbane), Magnetica Pty Ltd (MRI company based in Brisbane), alongside other industries interested in using low field MRI in their workflow. The Group has previously worked with Novartis Pharmaceutical (Australia) and atai Life Sciences (USA) in developing medical imaging methods to identify response to drugs. Some of their medical imaging technologies were developed in collaboration with Siemens Healthcare, and the Group continues to collaborate on projects.
Collaborations
Associate Professor Vegh has existing clinical collaborations with the Neurology Department of the Royal Brisbane and Women’s Hospital, and the Surgical Oncology Unit and the Medical Physics Department at the Princess Alexandra Hospital. These established collaborations allowed us to make connections with the Translational Research Institute (AI / machine learning; deep learning; radiomics, predictive modelling), Herston Imaging Research Facility (imaging data acquisition), and QIMR Berghofer (histology and genetics). These are existing links in Brisbane, Australia. The Group's low field MRI project involving exercise and professional sport is in collaboration with people from Human Movement Studies at the University of Queensland. Some of the AI and machine learning work is in collaboration with the School of Information Technology and Electrical Engineering, the University of Queensland.The Group's international collaborations include people from Harvard (USA) and the Materials and Physics Center (Spain) on low field MRI, and more widely on tissue microstructure imaging (UK, USA, and School of Mathematical Sciences, Queensland University of Technology). The latter is a consequence of Associate Professor Viktor Vegh leading the Anomalous Relaxation and Diffusion Study group, which brings together under one forum like-minded analytical people interesting in imaging tissue microstructure using MRI.
Funding
Large grants from the past few years which the Group held or continue to hold include: ARC Training Centre for Innovation in Biomedical Imaging Technology, Brisbane, Australia - developing of medical imaging technologies and training of the next generation of research technologists in collaboration with Siemens Healthcare. Cancer Council Queensland Project Grant – application of radiomics and machine learning in oesophageal cancer. BiomedTech Horizons 2.0 – low field MRI research and development to make MRI accessible, portable, and low-cost.
Key Publications
Su, Jiasheng and Pellicer-Guridi, Ruben and Edwards, Thomas and Fuentes, Miguel and Rosen, Matthew S and Vegh, Viktor and Reutens, David A CNN based software gradiometer for electromagnetic background noise reduction in low field MRI applications. IEEE Transactions on Medical Imaging 2022;41(5):1007-16
Vashistha, Rajat and Moradi, Hamed and Hammond, Amanda and O’Brien, Kieran and Rominger, Axel and Sari, Hasan and Shi, Kuangyu and Vegh, Viktor and Reutens, David ParaPET: non-invasive deep learning method for direct parametric brain PET reconstruction using histoimages. EJNMMI research 2024;14(1):10
Fard, Azin Shokraei and Reutens, David C and Vegh, Viktor From CNNs to GANs for cross-modality medical image estimation. Computers in Biology and Medicine 2022;146:105556
Sood, Surabhi and Urriola, Javier and Reutens, David and O’Brien, Kieran and Bollmann, Steffen and Barth, Markus and Vegh, Viktor Echo time-dependent quantitative susceptibility mapping contains information on tissue properties. Magnetic Resonance in Medicine 2017;77(5):1946-58
Vogel, Michael W and Guridi, Ruben Pellicer and Su, Jiasheng and Vegh, Viktor and Reutens, David C 3D-Spatial encoding with permanent magnets for ultra-low field magnetic resonance imaging. Scientific Reports 2019;9(1):1522