Abstract Nearly 40% of patients develop diabetes after an initial episode of acute pancreatitis (AP). Various studies have evaluated risk factors for the development of diabetes but they have shown inconsistent findings, suggesting methodologic shortcomings. The present proposal consists of three specific aims where we will attempt to determine the biochemical, radiologic and clinical factors related to the development of diabetes after AP. Aim 1: To identify autoantibodies associated with the progression of type 1 diabetes in patients after AP using a large human proteome array. Aim 2: To study the role of imaging, more specifically quantitative textural analysis to predict the development of type 1 diabetes after AP. Aim 3 is to build a machine learning model to predict type 1 diabetes after AP using patient-related risk factors, textural analysis on imaging and autoantibodies involved in disease progression. Our proposal will help us better understand diabetes after AP. Successful completion of this study has the potential to improve management for AP.