Explainability of a Deep Learning-Based Classification Model for Antineutrophil Cytoplasmic Autoantibody–Associated Glomerulonephritis

Introduction: The histopathological classification for antineutrophil cytoplasmic autoantibody (ANCA)–associated glomerulonephritis (ANCA-GN) is a well-established tool to reflect the variety of patterns and severity of lesions that can occur in kidney biopsies. It was demonstrated previously that d...

Full description

Saved in:
Bibliographic Details
Main Authors: Maria A.C. Wester Trejo, Maryam Sadeghi, Shivam Singh, Naghmeh Mahmoodian, Samir Sharifli, Zdenka Hruskova, Vladimír Tesař, Xavier Puéchal, Jan Anthonie Bruijn, Georg Göbel, Ingeborg M. Bajema, Andreas Kronbichler
Format: Article
Language:English
Published: Elsevier 2025-02-01
Series:Kidney International Reports
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2468024924020163
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849336192722534400
author Maria A.C. Wester Trejo
Maryam Sadeghi
Shivam Singh
Naghmeh Mahmoodian
Samir Sharifli
Zdenka Hruskova
Vladimír Tesař
Xavier Puéchal
Jan Anthonie Bruijn
Georg Göbel
Ingeborg M. Bajema
Andreas Kronbichler
author_facet Maria A.C. Wester Trejo
Maryam Sadeghi
Shivam Singh
Naghmeh Mahmoodian
Samir Sharifli
Zdenka Hruskova
Vladimír Tesař
Xavier Puéchal
Jan Anthonie Bruijn
Georg Göbel
Ingeborg M. Bajema
Andreas Kronbichler
author_sort Maria A.C. Wester Trejo
collection DOAJ
description Introduction: The histopathological classification for antineutrophil cytoplasmic autoantibody (ANCA)–associated glomerulonephritis (ANCA-GN) is a well-established tool to reflect the variety of patterns and severity of lesions that can occur in kidney biopsies. It was demonstrated previously that deep learning (DL) approaches can aid in identifying histopathological classes of kidney diseases; for example, of diabetic kidney disease. These models can potentially be used as decision support tools for kidney pathologists. Although they reach high prediction accuracies, their “black box” structure makes them nontransparent. Explainable (X) artificial intelligence (AI) techniques can be used to make the AI model decisions accessible for human experts. We have developed a DL-based model, which detects and classifies the glomerular lesions according to the Berden classification. Methods: Kidney biopsy slides of 80 patients with ANCA-GN from 3 European centers, who underwent a diagnostic kidney biopsy between 1991 and 2011, were included. We also investigated the explainability of our model using Gradient-weighted Class Activation Mapping (Grad-CAM) heatmaps. These maps were analyzed by pathologists to compare the decision-making criteria of humans and the DL model and assess the impact of different training settings. Results: The DL model shows a prediction accuracy of 93% for classifying lesions. The heatmaps from our trained DL models showed that the most predictive areas in the image correlated well with the areas deemed to be important by the pathologist. Conclusion: We present the first DL-based computational pipeline for classifying ANCA-GN kidney biopsies as per the Berden classification. XAI techniques helped us to make the decision-making criteria of the DL accessible for renal pathologists, potentially improving clinical decision-making.
format Article
id doaj-art-54c6dc91f14844fabc5ef194d5c46435
institution Kabale University
issn 2468-0249
language English
publishDate 2025-02-01
publisher Elsevier
record_format Article
series Kidney International Reports
spelling doaj-art-54c6dc91f14844fabc5ef194d5c464352025-08-20T03:45:03ZengElsevierKidney International Reports2468-02492025-02-0110245746510.1016/j.ekir.2024.11.005Explainability of a Deep Learning-Based Classification Model for Antineutrophil Cytoplasmic Autoantibody–Associated GlomerulonephritisMaria A.C. Wester Trejo0Maryam Sadeghi1Shivam Singh2Naghmeh Mahmoodian3Samir Sharifli4Zdenka Hruskova5Vladimír Tesař6Xavier Puéchal7Jan Anthonie Bruijn8Georg Göbel9Ingeborg M. Bajema10Andreas Kronbichler11Department of Pathology, Leiden University Medical Center, Leiden, The NetherlandsDepartment of Medical Statistics, Informatics and Health Economics, Medical University of Innsbruck, Innsbruck, AustriaDepartment of Medical Statistics, Informatics and Health Economics, Medical University of Innsbruck, Innsbruck, AustriaFaculty of Computer Science, Otto von Guericke University, Magdeburg, GermanyDepartment of Internal Medicine IV, Nephrology and Hypertension, Medical University of Innsbruck, Innsbruck, AustriaDepartment of Nephrology, First Faculty of Medicine, Charles University and General University Hospital in Prague, Prague, CzechiaDepartment of Nephrology, First Faculty of Medicine, Charles University and General University Hospital in Prague, Prague, CzechiaDepartment of Internal Medicine, Hôpital Cochin, AP-HP, Paris, FranceDepartment of Pathology, Leiden University Medical Center, Leiden, The NetherlandsDepartment of Medical Statistics, Informatics and Health Economics, Medical University of Innsbruck, Innsbruck, AustriaDepartment of Pathology and Medical Biology, University Medical Center Groningen, Groningen, The Netherlands; Correspondence: Ingeborg M. Bajema, Department of Pathology and Medical Biology, University Medical Center Groningen, Groningen, The Netherlands.Department of Internal Medicine IV, Nephrology and Hypertension, Medical University of Innsbruck, Innsbruck, Austria; Andreas Kronbichler, Department of Internal Medicine IV, Nephrology and Hypertension, Medical University of Innsbruck, Innsbruck, Austria.Introduction: The histopathological classification for antineutrophil cytoplasmic autoantibody (ANCA)–associated glomerulonephritis (ANCA-GN) is a well-established tool to reflect the variety of patterns and severity of lesions that can occur in kidney biopsies. It was demonstrated previously that deep learning (DL) approaches can aid in identifying histopathological classes of kidney diseases; for example, of diabetic kidney disease. These models can potentially be used as decision support tools for kidney pathologists. Although they reach high prediction accuracies, their “black box” structure makes them nontransparent. Explainable (X) artificial intelligence (AI) techniques can be used to make the AI model decisions accessible for human experts. We have developed a DL-based model, which detects and classifies the glomerular lesions according to the Berden classification. Methods: Kidney biopsy slides of 80 patients with ANCA-GN from 3 European centers, who underwent a diagnostic kidney biopsy between 1991 and 2011, were included. We also investigated the explainability of our model using Gradient-weighted Class Activation Mapping (Grad-CAM) heatmaps. These maps were analyzed by pathologists to compare the decision-making criteria of humans and the DL model and assess the impact of different training settings. Results: The DL model shows a prediction accuracy of 93% for classifying lesions. The heatmaps from our trained DL models showed that the most predictive areas in the image correlated well with the areas deemed to be important by the pathologist. Conclusion: We present the first DL-based computational pipeline for classifying ANCA-GN kidney biopsies as per the Berden classification. XAI techniques helped us to make the decision-making criteria of the DL accessible for renal pathologists, potentially improving clinical decision-making.http://www.sciencedirect.com/science/article/pii/S2468024924020163ANCAartificial intelligencehistopathologymachine learningvasculitis
spellingShingle Maria A.C. Wester Trejo
Maryam Sadeghi
Shivam Singh
Naghmeh Mahmoodian
Samir Sharifli
Zdenka Hruskova
Vladimír Tesař
Xavier Puéchal
Jan Anthonie Bruijn
Georg Göbel
Ingeborg M. Bajema
Andreas Kronbichler
Explainability of a Deep Learning-Based Classification Model for Antineutrophil Cytoplasmic Autoantibody–Associated Glomerulonephritis
Kidney International Reports
ANCA
artificial intelligence
histopathology
machine learning
vasculitis
title Explainability of a Deep Learning-Based Classification Model for Antineutrophil Cytoplasmic Autoantibody–Associated Glomerulonephritis
title_full Explainability of a Deep Learning-Based Classification Model for Antineutrophil Cytoplasmic Autoantibody–Associated Glomerulonephritis
title_fullStr Explainability of a Deep Learning-Based Classification Model for Antineutrophil Cytoplasmic Autoantibody–Associated Glomerulonephritis
title_full_unstemmed Explainability of a Deep Learning-Based Classification Model for Antineutrophil Cytoplasmic Autoantibody–Associated Glomerulonephritis
title_short Explainability of a Deep Learning-Based Classification Model for Antineutrophil Cytoplasmic Autoantibody–Associated Glomerulonephritis
title_sort explainability of a deep learning based classification model for antineutrophil cytoplasmic autoantibody associated glomerulonephritis
topic ANCA
artificial intelligence
histopathology
machine learning
vasculitis
url http://www.sciencedirect.com/science/article/pii/S2468024924020163
work_keys_str_mv AT mariaacwestertrejo explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT maryamsadeghi explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT shivamsingh explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT naghmehmahmoodian explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT samirsharifli explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT zdenkahruskova explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT vladimirtesar explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT xavierpuechal explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT jananthoniebruijn explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT georggobel explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT ingeborgmbajema explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis
AT andreaskronbichler explainabilityofadeeplearningbasedclassificationmodelforantineutrophilcytoplasmicautoantibodyassociatedglomerulonephritis