Computational Renal Pathology

Correlating Reference Kidney Morphometry with Patient Demographics & Creatinine

We leverage the unique benefits of panoptic segmentation to perform the largest ever quantitation of reference kidney morphometry. Kidney features were found to vary with age and sex; and glomeruli size may intricately link to creatinine, defying prior notions.


PodoCount: Labeled Podocyte Quantification Tool

We have developed PodoCount, an automated tool for podocyte detection and three-dimensional quantification from histologic sections of renal tissue biopsy whole slide images (WSIs) with podocytes labeled using immunohistochemistry labels.


PodoSighter: Label-Free Podocyte Detection

We have developed PodoSighter, a cloud-based tool for automated, label-free podocyte detection and three-dimensional quantification from periodic acid-Schiff-stained histologic sections of renal tissue biopsy WSIs.


Computational Segmentation and Quantification of Interstital Fibrosis, Tubular Atrophy, & Glomerulosclerosis

A machine learning (ML) pipeline is developed to conduct reliable, digital, and automated detection of interstitial fibrosis and tubular atrophy (IFTA).


Computational Segmentation and Classification of Diabetic Glomerulosclerosis

We developed a digital pipeline to classify renal biopsies from patients with diabetic nephropathy (DN). We combined traditional image analysis with modern machine learning to efficiently capture important structures, minimize manual effort and supervision, and enforce biologic prior information onto our model.


Supervised Computational Detection of Glomeruli in Renal Biopsies

We have developed a supervised computational pipeline to detect glomeruli from digital renal biopsy WSIs.


Unsupervised Computational Labeling of Glomerular Textural Boundaries in Renal Pathology

We have developed an unsupervised automated quantification of glomerulus from digital renal biopsy WSIs.