Assistant Professor Brigham and Women's Hospital Boston, Massachusetts, United States
Description: This session delves into the rapidly emerging field of computational pathology, emphasizing the development of objective prognostic models from histology images and genomics. Participants will explore how multimodal deep learning integrates pathology whole-slide images and molecular profile data from different cancer types to create joint image-omic prognostic models. The session will highlight the discovery of explainable morphological and molecular descriptors that govern prognosis, with all analyses presented in an interactive open-access database for further exploration and biomarker discovery.
Learning Objectives: 1. Understand how multimodal data fusion improves prognostic models for various cancer types. 2. Comprehend the use of multimodal deep learning to merge pathology whole-slide images and molecular profile data.
