Nassim Bouteldja1, Barbara M Klinkhammer2,3, Roman D Bülow2, Patrick Droste2, Simon W Otten2, Saskia Freifrau von Stillfried2, Julia Moellmann4, Susan M Sheehan5, Ron Korstanje5, Sylvia Menzel3, Peter Bankhead6,7, Matthias Mietsch8, Charis Drummer9, Michael Lehrke4, Rafael Kramann3,10, Jürgen Floege3, Peter Boor11,3, Dorit Merhof1,12. 1. Institute of Imaging and Computer Vision, RWTH Aachen University, Aachen, Germany. 2. Institute of Pathology, RWTH Aachen University Hospital, Aachen, Germany. 3. Department of Nephrology and Immunology, RWTH Aachen University Hospital, Aachen, Germany. 4. Department of Cardiology and Vascular Medicine, RWTH Aachen University Hospital, Aachen, Germany. 5. The Jackson Laboratory, Bar Harbor, Maine. 6. Edinburgh Pathology, University of Edinburgh, Edinburgh, United Kingdom. 7. Institute of Genetics and Molecular Medicine, University of Edinburgh, Edinburgh, United Kingdom. 8. Laboratory Animal Science Unit, German Primate Center, Goettingen, Germany. 9. Platform Degenerative Diseases, German Primate Center, Goettingen, Germany. 10. Department of Internal Medicine, Nephrology and Transplantation, Erasmus Medical Center, Rotterdam, The Netherlands. 11. Institute of Pathology, RWTH Aachen University Hospital, Aachen, Germany pboor@ukaachen.de. 12. Fraunhofer Institute for Digital Medicine MEVIS, Bremen, Germany.
Abstract
BACKGROUND: Nephropathologic analyses provide important outcomes-related data in experiments with the animal models that are essential for understanding kidney disease pathophysiology. Precision medicine increases the demand for quantitative, unbiased, reproducible, and efficient histopathologic analyses, which will require novel high-throughput tools. A deep learning technique, the convolutional neural network, is increasingly applied in pathology because of its high performance in tasks like histology segmentation. METHODS: We investigated use of a convolutional neural network architecture for accurate segmentation of periodic acid-Schiff-stained kidney tissue from healthy mice and five murine disease models and from other species used in preclinical research. We trained the convolutional neural network to segment six major renal structures: glomerular tuft, glomerulus including Bowman's capsule, tubules, arteries, arterial lumina, and veins. To achieve high accuracy, we performed a large number of expert-based annotations, 72,722 in total. RESULTS: Multiclass segmentation performance was very high in all disease models. The convolutional neural network allowed high-throughput and large-scale, quantitative and comparative analyses of various models. In disease models, computational feature extraction revealed interstitial expansion, tubular dilation and atrophy, and glomerular size variability. Validation showed a high correlation of findings with current standard morphometric analysis. The convolutional neural network also showed high performance in other species used in research-including rats, pigs, bears, and marmosets-as well as in humans, providing a translational bridge between preclinical and clinical studies. CONCLUSIONS: We developed a deep learning algorithm for accurate multiclass segmentation of digital whole-slide images of periodic acid-Schiff-stained kidneys from various species and renal disease models. This enables reproducible quantitative histopathologic analyses in preclinical models that also might be applicable to clinical studies.
BACKGROUND: Nephropathologic analyses provide important outcomes-related data in experiments with the animal models that are essential for understanding kidney disease pathophysiology. Precision medicine increases the demand for quantitative, unbiased, reproducible, and efficient histopathologic analyses, which will require novel high-throughput tools. A deep learning technique, the convolutional neural network, is increasingly applied in pathology because of its high performance in tasks like histology segmentation. METHODS: We investigated use of a convolutional neural network architecture for accurate segmentation of periodic acid-Schiff-stained kidney tissue from healthy mice and five murine disease models and from other species used in preclinical research. We trained the convolutional neural network to segment six major renal structures: glomerular tuft, glomerulus including Bowman's capsule, tubules, arteries, arterial lumina, and veins. To achieve high accuracy, we performed a large number of expert-based annotations, 72,722 in total. RESULTS: Multiclass segmentation performance was very high in all disease models. The convolutional neural network allowed high-throughput and large-scale, quantitative and comparative analyses of various models. In disease models, computational feature extraction revealed interstitial expansion, tubular dilation and atrophy, and glomerular size variability. Validation showed a high correlation of findings with current standard morphometric analysis. The convolutional neural network also showed high performance in other species used in research-including rats, pigs, bears, and marmosets-as well as in humans, providing a translational bridge between preclinical and clinical studies. CONCLUSIONS: We developed a deep learning algorithm for accurate multiclass segmentation of digital whole-slide images of periodic acid-Schiff-stained kidneys from various species and renal disease models. This enables reproducible quantitative histopathologic analyses in preclinical models that also might be applicable to clinical studies.
Authors: Meyke Hermsen; Thomas de Bel; Marjolijn den Boer; Eric J Steenbergen; Jesper Kers; Sandrine Florquin; Joris J T H Roelofs; Mark D Stegall; Mariam P Alexander; Byron H Smith; Bart Smeets; Luuk B Hilbrands; Jeroen A W M van der Laak Journal: J Am Soc Nephrol Date: 2019-09-05 Impact factor: 10.121
Authors: Brendon Lutnick; Brandon Ginley; Darshana Govind; Sean D McGarry; Peter S LaViolette; Rabi Yacoub; Sanjay Jain; John E Tomaszewski; Kuang-Yu Jen; Pinaki Sarder Journal: Nat Mach Intell Date: 2019-02-11
Authors: Korsuk Sirinukunwattana; Shan E Ahmed Raza; David R J Snead; Ian A Cree; Nasir M Rajpoot Journal: IEEE Trans Med Imaging Date: 2016-02-04 Impact factor: 10.048
Authors: Maike Baues; Barbara M Klinkhammer; Josef Ehling; Felix Gremse; Marc A M J van Zandvoort; Chris P M Reutelingsperger; Christoph Daniel; Kerstin Amann; Janka Bábíčková; Fabian Kiessling; Jürgen Floege; Twan Lammers; Peter Boor Journal: Kidney Int Date: 2019-09-18 Impact factor: 10.612
Authors: Barbara Mara Klinkhammer; Sonja Djudjaj; Uta Kunter; Runolfur Palsson; Vidar Orn Edvardsson; Thorsten Wiech; Margret Thorsteinsdottir; Sverrir Hardarson; Orestes Foresto-Neto; Shrikant R Mulay; Marcus Johannes Moeller; Wilhelm Jahnen-Dechent; Jürgen Floege; Hans-Joachim Anders; Peter Boor Journal: J Am Soc Nephrol Date: 2020-02-21 Impact factor: 10.121
Authors: Peter Bankhead; Maurice B Loughrey; José A Fernández; Yvonne Dombrowski; Darragh G McArt; Philip D Dunne; Stephen McQuaid; Ronan T Gray; Liam J Murray; Helen G Coleman; Jacqueline A James; Manuel Salto-Tellez; Peter W Hamilton Journal: Sci Rep Date: 2017-12-04 Impact factor: 4.379
Authors: Elise Marechal; Adrien Jaugey; Georges Tarris; Michel Paindavoine; Jean Seibel; Laurent Martin; Mathilde Funes de la Vega; Thomas Crepin; Didier Ducloux; Gilbert Zanetta; Sophie Felix; Pierre Henri Bonnot; Florian Bardet; Luc Cormier; Jean-Michel Rebibou; Mathieu Legendre Journal: Clin J Am Soc Nephrol Date: 2021-12-03 Impact factor: 8.237
Authors: Roman D Büllow; Jon N Marsh; S Joshua Swamidass; Joseph P Gaut; Peter Boor Journal: Curr Opin Nephrol Hypertens Date: 2022-02-14 Impact factor: 3.416
Authors: Xiang Li; Richard C Davis; Yuemei Xu; Zehan Wang; Nao Souma; Gina Sotolongo; Jonathan Bell; Matthew Ellis; David Howell; Xiling Shen; Kyle J Lafata; Laura Barisoni Journal: J Med Imaging (Bellingham) Date: 2021-12-20
Authors: Yi Zheng; Clarissa A Cassol; Saemi Jung; Divya Veerapaneni; Vipul C Chitalia; Kevin Y M Ren; Shubha S Bellur; Peter Boor; Laura M Barisoni; Sushrut S Waikar; Margrit Betke; Vijaya B Kolachalama Journal: Am J Pathol Date: 2021-05-23 Impact factor: 5.770