Artificial intelligence and the interplay between tumor and immunity

Digital pathology image analysis and deep learning can be utilized to quantify and characterize nuanced interactions between cancer and the immune system. Recent advances in deep learning and artificial intelligence in Pathomics data have led to the development of methods and techniques that augment and empower qualitative traditional diagnostic histopathologic evaluation in order to substantially accelerate cancer research. Emerging digital pathology and deep learning applications can (1) stratify patient management through data-driven insights into cancer, (2) identify relevant biomarkers to predict clinical outcomes and treatment response, (3) enhance our collective understanding of cancer biology to motivate the utilization of novel therapeutic approaches. This chapter introduces and describes a selected set of novel Pathomics-based deep learning methods that have been developed to classify and reproducibly quantify the interplay between tumor cells and the immune response in the tumor microenvironment.