webinar register page

Webinar banner
Developing AI Models for Medical Image Analysis with Efficient and Low-cost Annotations
The fast development of AI techniques has revolutionized medical image analysis for computer-assisted diagnosis and treatment systems in recent years. The success of deep learning relies highly on a large set of annotated images for training. However, annotating medical images is expensive and time-consuming, due to the large size and low contrast of medical images and limited access to experts for accurate annotation. This talk will introduce several techniques of reducing the annotation cost for developing high-performance models, especially for medical image segmentation. I will present our recent works on efficient interactive annotation, weakly supervised learning, semi-supervised learning, domain adaptation and noisy label learning for medical image segmentation. These methods have been applied to a range of tasks, such as segmentation of brain tumors, covid-19 lesions, vessels and abdominal organs.

Sep 2, 2022 05:00 PM in Hong Kong SAR

* Required information


Prof. Guotai Wang
@University of Electronic Science and Technology of China
Dr. Guotai Wang is an Associate Professor at the School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China. He received his PhD degree at the Centre for Medical Image Computing, University College London in 2018. His research interests are computer vision, deep learning and medical image analysis. He has 40+ publications in IEEE TPAMI, IEEE TMI, MedIA, AAAI, MICCAI, etc. He serves as Associate Editor of Medical Physics and guest editor of MedIA in 2022, and served as Area Chair of MICCAI 2021 and ISBI 2021. He was also recognized as an outstanding reviewer for IEEE TMI and MedIA. In recent years, he led students winning several MICCAI challenges, including EndoVis 2019 and MyoPS 2020.