연사: 전세영, 서울대학교 교수
일시: 2021년 4월 1일 (목), 17:00 ~ 18:00

장소: 화상강의

Abstract:

Medical imaging domains have resorted to advanced computations not only for achieving the best performance, but also for attaining the best reliability and safety in their mission-critical applications (e.g., saving lives) for a long time. They pose challenging restrictions in medical imaging such as ALARA (as low as reasonably achievable) that could be often overcome by intelligent computations. In this talk, firstly, I will share my research experience on cancer imaging (positron emission tomography), starting from theoretical works to applications. Then, I will talk about my two recent projects on deep learning potentially for medical imaging applications: (1) unsupervised learning for image recovery and (2) active learning for image classification / semantic segmentation. Finally, if time permits, I will briefly share my future research plans at Intelligent Computational imaging Lab (ICL), Department of Electrical and Computer Engineering, Seoul National University.

Biography:

l 학력

경력