세미나 제목: Towards Robust Multi-site Neuroimaging Analyses and Applications

세미나 초록:Modern neuroimaging studies often combine data from multiple scanners and acquisition protocols. Such collection of data may contain substantial technical variability associated with scanner effects where images from different scanners possess non-biological, systematic variability. Interestingly, from a machine learning perspective, such multi-site/scanner neuroimaging data also poses an empirical challenge: a model trained on one scanner performs poorly when tested on another scanner. On the application side, I will discuss our recent works by first characterizing this problem with a pre-existing notion of domain generalization from computer vision. Then, I will show how a domain generalization technique can significantly improve the generalizability of a brain lesion detection model on new, unseen neuroimaging data. On the analysis side, I will also show some preliminary results from our recent investigation on neuroimaging data harmonization to reduce scanner variability across scanners on a unique local dataset

연사 이력: Seong Jae Hwang received his B.S. in Computer Science from the University of Illinois at Urbana-Champaign in 2011, M.S.E. in Robotics from the University of Pennsylvania in 2013, and Ph.D. in Computer Sciences from the University of Wisconsin-Madison in 2019. He was an assistant professor of Computer Science and Intelligent Systems Program in the School of Computing and Information at the University of Pittsburgh from 2019 to 2021, before joining the Department of Artificial Intelligence at Yonsei University as an assistant professor in 2022

연사 정보: https://sites.pitt.edu/~sjh95/