Abstract: In this talk, I will discuss the challenges and opportunities of developing “production-ready” machine learning models and deploying them into clinical practice. I will discuss a number of ML projects I have been involved in, including body composition analysis in CT, adrenal mass detection and classification in CT, and stroke detection and delineation in CT and MRI. But I will also discuss my experience with deploying models into clinical practice, the practical steps necessary, and the challenges that we are yet to solve.

About the speaker: Christopher Bridge, DPhil completed his BA and MEng in Computer and Information Engineering at the University of Cambridge and his DPhil at the University of Oxford, where he focused on the application of machine learning to analysis of fetal heart cardiac videos. From 2017 to 2022, he worked as a machine learning scientist and later as director of machine learning at the Mass General Brigham Data Science Office (DSO). At the DSO he focused on the development of commercial machine learning models with industry partnerships, as well as on the deployment of machine learning models into clinical practice. He is now an Instructor in Radiology in the Quantitative Translational Imaging in Medicine (QTIM) laboratory at the MGH’s Athinoula A. Martinos Center for Biomedical Imaging. His research interests span applications of machine learning across radiology and pathology imaging, ML for precision oncology, and open-source software and standards for clinical deployment of ML models.