Project Grant R01DK149216

The National Institute of Diabetes and Digestive and Kidney Diseases awarded the Trustees of Boston University $791,614 on September 1, 2026, under the Diabetes, Digestive, and Kidney Diseases Extramural Research program (CFDA 93.847) to develop an artificial intelligence–driven digital pathology platform for kidney disease diagnosis and research.

The recipient will build foundational deep learning models capable of automated segmentation, quantification, and disease classification from transmission electron microscopy (TEM) images of the kidney to streamline clinical workflows and increase diagnostic accuracy. The team will create a multimodal integration framework combining data from routine histology, spatial transcriptomics, and TEM images using contrastive and advanced deep learning techniques to enable more precise disease prediction and patient stratification. The project will also develop virtual staining tools to predict TEM and immunofluorescence findings directly from standard histology images, extending diagnostic reach to hospitals and research centers with limited access to advanced microscopy.

The work addresses the shortage of trained renal pathologists and labor-intensive TEM analysis that constrains timely diagnosis and treatment of glomerular diseases, a frequent cause of chronic kidney disease. The recipient organization is Boston University Medical Campus, located in Roxbury, Massachusetts. The project grant runs through June 30, 2030.