Computational Image Analysis Platform (CIMAP) for HuBMAP: Mapping OMICS to Brightfield Histology

Abstract for team project led by Dr. Sarder: The proposed work is for the Tools component of the HuBMAP consortium. The purpose of the work is to provide information about histological tissue organization at the cellular level that can be used to interpret molecular data from HuBMAP and infer molecular states from histological information. The current proposal is for one year. The project will use images of frozen sections of kidney tissue which show both tissue morphology and localizations of specific types of molecules. The team will use machine learning to cluster the images into groups that are meaningful for understanding the molecular basis of disease, with the goal of developing algorithms that will classify tissue sections.