Developing a deep learning model in the electronic health record database to predict colorectal cancer risk after colonoscopy screening
This project aims to extract and process detailed electronic health record data for the future development of a machine learning-based model for personalized colonoscopy surveillance in a large integrated healthcare system. We will seek further funding to develop a risk assessment tool for the prediction of advanced neoplasia among adults with a history of polyp removal. Leveraging our unique resources and experienced transdisciplinary team, the proposed research represents the next critical step in harnessing the great power of data science for personalized cancer care and has significant implications for improving healthcare delivery to better prevent colorectal cancer.