Abstract: Magnetic resonance (MR) imaging is the reference imaging modality both for functional and structural study of the human brain. Diffusion MR imaging allows measuring the water diffusion properties in every voxel of a volume. When considering the white matter of the brain, this low-level information allows to infer higher-level information, namely about the orientation of the neuronal fiber bundles, and therefore about brain connectivity as a whole. In this talk we will first give a brief introduction to the physical phenomenon of diffusion and how it can be used in MRI to infer brain neuronal fiber orientation information. Then we will show that it actually can provide much more information than just fiber orientation. Indeed, by developing realistic tissue microstructure models, advanced diffusion MR sequences and robust model estimation techniques, we will show how to estimate tissue parameters at the microscopic scale from the macroscopic MR signals. In particular, we will introduce some of our recent works in this domain, where hyper-realistic synthetic tissue models, advanced Monte-Carlo simulations and ML-based estimation techniques are developed and combined. We will show the application of those techniques to address the analysis of the brain microstructure and discuss their potential in the analysis of other tissues.
Short bio: Jean-Philippe Thiran was born in Namur, Belgium, in 1970. He received the Electrical Engineering degree and the PhD degree from the Université catholique de Louvain (UCL), Louvain-la-Neuve, Belgium, in 1993 and 1997, respectively. He joined the Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland in 1998. Dr Thiran is currently a Full Professor at EPFL, Director of the Signal Processing Laboratory (LTS5). Moreover, he currently serves as the Director of the Institute of Electrical and Micro Engineering at EPFL. His research field is computational imaging, with applications in many domains including medical image analysis (diffusion MRI, ultrasound imaging and digital pathology) and computer vision. He also holds a part-time Associate Professor position with the Department of Radiology of the University Hospital Center (CHUV) and University of Lausanne (UNIL). Prof. Thiran is author or co-author of 1 book, 9 book chapters, 250 journal papers and more than 270 peer-reviewed papers published in the proceedings of international conferences. He holds 12 international patents. From 2001 to 2005 he was Co-Editor-in-Chief of the Signal Processing international journal (published by Elsevier Science). He has been an associate editor of the IEEE Transactions on Image Processing. Among many other duties, he has been the general chairman of the 2008 European Signal Processing Conference (EUSIPCO 2008) and the Technical Co-chair of the 2015 IEEE International Conference on Image Processing (IEEE ICIP 2015). He is a senior member of the IEEE and a Fellow of the European Association for Signal Processing (EURASIP). Previous webinar videos are available online at https://vimeo.com/showcase/8005816