Deep-Learning based segmentation and quantification in experimental kidney histopathology

Nassim Bouteldja, Barbara M Klinkhammer, Roman D. Bulow, Patrick Droste, Simon W. Otten, Saskia von Stillfried, Julia Moellmann, Susan M Sheehan, Ron Korstanje, Sylvia Menzel, Peter Bankhead, Matthias Mietsch, Charis Drummer, Michael Lehrke, Rafael Kramann, Jurgen Floege, Peter Boor, Dorit Merhof

Research output: Contribution to journalArticlepeer-review

Abstract

Preclinical animal experiments are of high importance in nephrology research, with histology as a major readout. Here, the authors provide a multiclass histology segmentation tool to evaluate animal kidney disease models using deep learning. A convolutional neural network (CNN) enabled a rapid, automated, high-performance whole slide segmentation of renal histology, allowing high-throughput analyses in various species and multiple murine disease models. The CNN also showed high performance in patient samples, providing a translational bridge between preclinical and clinical research. Extracted quantitative morphological features closely correlated with standard morphometric measurements. In conclusion, deep learning-based segmentation in experimental renal pathology opens new dimensions of reproducible,
unbiased and high-throughput quantitative digital nephropathology.
Original languageEnglish
JournalJournal of the American Society of Nephrology
Early online date5 Nov 2020
DOIs
Publication statusE-pub ahead of print - 5 Nov 2020

Fingerprint Dive into the research topics of 'Deep-Learning based segmentation and quantification in experimental kidney histopathology'. Together they form a unique fingerprint.

Cite this