Assessment of glomerular morphological patterns by deep learning algorithms

Marlen Runz,Zoran V. Popovic,Stefan Porubsky,Jan Niklas Bindzus,Cleo-Aron Weis,Matthias M. Gaida,Jonas Voigt,Svetlana Hertjens

doi:10.1007/s40620-021-01221-9

Abstract

BackgroundCompilation of different morphological lesion signatures is characteristic of renal pathology. Previous studies have documented the potential value of artificial intelligence (AI) in recognizing relatively clear-cut glomerular structures and patterns, such as segmental or global sclerosis or mesangial hypercellularity. This study aimed to test the capacity of deep learning algorithms to recognize complex glomerular structural changes that reflect common diagnostic dilemmas in nephropathology.MethodsFor this purpose, we defined nine classes of glomerular morphological patterns and trained twelve convolutional neuronal network (CNN) models on these. The two-step training process was done on a first dataset defined by an expert nephropathologist (12,253 images) and a second consensus dataset (11,142 images) defined by three experts in the field.ResultsThe efficacy of CNN training was evaluated using another set with 180 consensus images, showing convincingly good classification results (kappa-values 0.838–0.938).Furthermore, we elucidated the image areas decisive for CNN-based decision making by class activation maps. Finally, we demonstrated that the algorithm could decipher glomerular disease patterns coinciding in a single glomerulus (e.g. necrosis along with mesangial and endocapillary hypercellularity).ConclusionsIn summary, our model, focusing on glomerular lesions detectable by conventional microscopy, is the first sui generis to deploy deep learning as a reliable and promising tool in recognition of even discrete and/or overlapping morphological changes. Our results provide a stimulus for ongoing projects that integrate further input levels next to morphology (such as immunohistochemistry, electron microscopy, and clinical information) to develop a novel tool applicable for routine diagnostic nephropathology.

Highlights

MethodsArtificial intelligence is an umbrella term encompassing techniques that enable machines to mimic human intelligence
We investigated the potential value of machine learning in automatized recognition of common glomerular morphological changes
We initially intended to exclude the influence of interobserver variability and focus on the capacity of convolutional neuronal network (CNN) to recognize clear-cut patterns

Summary

Introduction

MethodsArtificial intelligence is an umbrella term encompassing techniques that enable machines to mimic human intelligence. Due to the plethora of sometimes overlapping clinical and histological disease patterns, nephropathology represents a challenging discipline. This holds true for the evaluation of histology sections from experiments with kidney diseases in animal models. Previous studies have documented the potential value of artificial intelligence (AI) in recognizing relatively clear-cut glomerular structures and patterns, such as segmental or global sclerosis or mesangial hypercellularity. Methods For this purpose, we defined nine classes of glomerular morphological patterns and trained twelve convolutional neuronal network (CNN) models on these. Conclusions In summary, our model, focusing on glomerular lesions detectable by conventional microscopy, is the first sui generis to deploy deep learning as a reliable and promising tool in recognition of even discrete and/or overlapping morphological changes. Our results provide a stimulus for ongoing projects that integrate further input levels next to morphology (such as immunohistochemistry, electron microscopy, and clinical information) to develop a novel tool applicable for routine diagnostic nephropathology

Objectives

Results

Conclusion