Doctoral Thesis: Learning Representations for Limited and Heterogeneous Medical Data
Data insufficiency and heterogeneity are challenges of representation learning for machine learning in medicine due to the diversity of medical data and the expense of data collection and annotation. To learn generalizable representations from such limited and heterogeneous medical data, we aim to utilize various learning paradigms to overcome the issue. In the talk, we systematically explore the machine learning frameworks for limited data, data imbalance, and heterogeneous data, using cross-domain learning, self-supervised learning, and meta-learning. We present studies with different medical applications, such as clinical language translation, ultrasound image classification and segmentation, skin diagnosis classification. The investigation in this talk is not exhaustive but it introduces an extensive understanding of utilizing machine learning in helping clinical decision making under the limited and heterogeneous medical data setting.
Thesis Supervisor: Prof. Peter Szolovits
Zoom link: https://mit.zoom.us/my/ckbjimmy
- Date: Tuesday, April 5
- Time: 1:30 pm - 5:00 pm
- Location: 32-G449