Fight for Fairness in AI for Medical Imaging
- 2023-09-11 (Mon.), 10:30 AM
- 統計所B1演講廳;茶 會:上午10:10。
- 中文演講,實體與線上視訊同步進行。
- Prof. Po-Chih (Bruce) Kuo ( 郭柏志 教授 )
- 國立清華大學資訊工程學系
Abstract
Deep learning models have shown promise in medical imaging analysis, but concerns exist regarding biases leading to performance disparities among protected groups (e.g., age, race, gender). It has been shown that AI models can learn race on medical images, leading to algorithmic bias. In this talk, I will go through our recent approaches to enhance the fairness of medical image models by eliminating bias related to race, age, and gender. We hypothesize models may be learning patients' demographics via shortcut learning and combat this using self-supervised learning, image augmentation, siamese networks and adversarial learning.
線上視訊請點選連結
線上視訊請點選連結
附件下載
最後更新日期:2023-08-28 09:06