The overall objective of this project is to develop a deep learning based imaging analysis and informatics tool for better understanding of pathogenic mechanisms in T1D, using existing data which has been collected for 14 years by Network for Pancreatic Organ Donors with Diabetes (nPOD) study. The central hypothesis of this proposal is that modifications of the pancreatic islets due to loss of insulin-producing β-cells and infiltration of islets by CD3+T cells during the progression to T1D progression can be characterized and their heterogeneity can be quantified in pancreas whole-slide digital pathology images. Ultimately, our results will help in the rational design of T1D prevention and treatment strategies.
