Integrating Pathology Images and Genomics Data for Cancer Grading
Explore a 48-minute conference talk by Dr. Xiaohan Xing from Stanford University on integrating pathology images and genomics data for cancer grading. Learn about innovative AI techniques addressing challenges in multi-modal biomedical data analysis, including a low-rank constraint-based method for bridging modality gaps, a saliency-aware masking strategy for balancing modal contributions, and a knowledge distillation framework for handling missing genomics data. Discover how these approaches were validated using the TCGA GBMLGG dataset, enhancing cancer grading accuracy and reliability. Gain insights from Dr. Xing's extensive research in medical AI, focusing on disease diagnosis and survival prediction using medical images and omics data integration.