Secretariat, Alumni Association, IDAC
Date Wednesday, 1 March 2023, 13:30~14:30
Room Online (Zoom)
Title Diffusion MRI by Machine Learning
Speaker Yoshitaka Masutani Ph.D
Affiliation Medical Image Computation Lab, Division of Health Sciences, Tohoku University Graduate School of Medicine
Organizer Yasuyuki Taki (Dept. of Aging Research and Geriatric Medicine・ext 8559)
Yasuko Tatewaki (Dept. of Aging Research and Geriatric Medicine・ext 8559)
Abstract In this talk, the basics and recent developments in quantitative acquisition of the structural and functional features in the living bodies by diffusion MRI (dMRI) are presented. First, various signal models such as DTI, DKI, and NODDI with their classification from the viewpoint of parameter inference methods are introduced. Next, mathematical generalizations and methodologies in the parameter inference are presented including the development from conventional
fitting to recent machine learning approaches. In addition, an original technique called synthetic Q-space learning is described with its advantages and disadvantages by showing application results in various signal models. Finally, future prospects for dMRI parameter inference are discussed from the aspects of imaging, modeling and simulation. The main content of this seminar is available in a review paper by the speaker [1].
[1] Masutani Y, Recent Advances in Parameter Inference for Diffusion MRI Signal Models, Magn Reson Med Sci 21(1): 132-147, 2022