This project aims to address the problem of machine learning model deterioration due to distribution shifts under clinical settings. For example, when a model is deployed in another hospital with different clinical practices, disease trends, or electronic health record coding systems, it may lead to false diagnostics in clinical support tools that adversely affect patient safety.

I am participating in SuperUROP because I want to gain more research experience through a long-term research project in the field of clinical machine learning. I look forward to furthering my knowledge and research skills, as well as hopefully making a positive contribution to this field. I’ m excited to work on this project because it has real-world applications that can enhance the reliability of healthcare machine learning systems.