A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer

The Gleason score is an important predictor of prognosis in prostate cancer. However, its subjective nature can result in over- or under-grading. Our objective was to train an artificial intelligence (AI)-based algorithm to grade prostate cancer in specimens from patients who underwent radical prost...

Full description

Saved in:
Bibliographic Details
Main Authors: Sushant Patkar, Stephanie Harmon, Isabell Sesterhenn, Rosina Lis, Maria Merino, Denise Young, G. Thomas Brown, Kimberly M. Greenfield, John D. McGeeney, Sally Elsamanoudi, Shyh-Han Tan, Cara Schafer, Jiji Jiang, Gyorgy Petrovics, Albert Dobi, Francisco J. Rentas, Peter A. Pinto, Gregory T. Chesnut, Peter Choyke, Baris Turkbey, Joel T. Moncur
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:Journal of Pathology Informatics
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2153353924000208
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846122140821094400
author Sushant Patkar
Stephanie Harmon
Isabell Sesterhenn
Rosina Lis
Maria Merino
Denise Young
G. Thomas Brown
Kimberly M. Greenfield
John D. McGeeney
Sally Elsamanoudi
Shyh-Han Tan
Cara Schafer
Jiji Jiang
Gyorgy Petrovics
Albert Dobi
Francisco J. Rentas
Peter A. Pinto
Gregory T. Chesnut
Peter Choyke
Baris Turkbey
Joel T. Moncur
author_facet Sushant Patkar
Stephanie Harmon
Isabell Sesterhenn
Rosina Lis
Maria Merino
Denise Young
G. Thomas Brown
Kimberly M. Greenfield
John D. McGeeney
Sally Elsamanoudi
Shyh-Han Tan
Cara Schafer
Jiji Jiang
Gyorgy Petrovics
Albert Dobi
Francisco J. Rentas
Peter A. Pinto
Gregory T. Chesnut
Peter Choyke
Baris Turkbey
Joel T. Moncur
author_sort Sushant Patkar
collection DOAJ
description The Gleason score is an important predictor of prognosis in prostate cancer. However, its subjective nature can result in over- or under-grading. Our objective was to train an artificial intelligence (AI)-based algorithm to grade prostate cancer in specimens from patients who underwent radical prostatectomy (RP) and to assess the correlation of AI-estimated proportions of different Gleason patterns with biochemical recurrence-free survival (RFS), metastasis-free survival (MFS), and overall survival (OS). Training and validation of algorithms for cancer detection and grading were completed with three large datasets containing a total of 580 whole-mount prostate slides from 191 RP patients at two centers and 6218 annotated needle biopsy slides from the publicly available Prostate Cancer Grading Assessment dataset. A cancer detection model was trained using MobileNetV3 on 0.5 mm × 0.5 mm cancer areas (tiles) captured at 10× magnification. For cancer grading, a Gleason pattern detector was trained on tiles using a ResNet50 convolutional neural network and a selective CutMix training strategy involving a mixture of real and artificial examples. This strategy resulted in improved model generalizability in the test set compared with three different control experiments when evaluated on both needle biopsy slides and whole-mount prostate slides from different centers. In an additional test cohort of RP patients who were clinically followed over 30 years, quantitative Gleason pattern AI estimates achieved concordance indexes of 0.69, 0.72, and 0.64 for predicting RFS, MFS, and OS times, outperforming the control experiments and International Society of Urological Pathology system (ISUP) grading by pathologists. Finally, unsupervised clustering of test RP patient specimens into low-, medium-, and high-risk groups based on AI-estimated proportions of each Gleason pattern resulted in significantly improved RFS and MFS stratification compared with ISUP grading. In summary, deep learning-based quantitative Gleason scoring using a selective CutMix training strategy may improve prognostication after prostate cancer surgery.
format Article
id doaj-art-d4d03e7f2d064afd82948e3605766bf5
institution Kabale University
issn 2153-3539
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series Journal of Pathology Informatics
spelling doaj-art-d4d03e7f2d064afd82948e3605766bf52024-12-15T06:15:16ZengElsevierJournal of Pathology Informatics2153-35392024-12-0115100381A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancerSushant Patkar0Stephanie Harmon1Isabell Sesterhenn2Rosina Lis3Maria Merino4Denise Young5G. Thomas Brown6Kimberly M. Greenfield7John D. McGeeney8Sally Elsamanoudi9Shyh-Han Tan10Cara Schafer11Jiji Jiang12Gyorgy Petrovics13Albert Dobi14Francisco J. Rentas15Peter A. Pinto16Gregory T. Chesnut17Peter Choyke18Baris Turkbey19Joel T. Moncur20Artificial Intelligence Resource, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USAArtificial Intelligence Resource, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA; Corresponding author at: 9000 Rockville Pike, Building 10, Room B3B69F, Artificial Intelligence Resource, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA.The Joint Pathology Center, Silver Spring, MD 20910, USAArtificial Intelligence Resource, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USALaboratory of Pathology, National Cancer Institute, National Institutes of Health, Bethesda, MD, USACenter for Prostate Disease Research, Murtha Cancer Center Research Program, Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD 20817, USA; Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., Bethesda, MD 20817, USAArtificial Intelligence Resource, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USAThe Joint Pathology Center, Silver Spring, MD 20910, USAThe Joint Pathology Center, Silver Spring, MD 20910, USACenter for Prostate Disease Research, Murtha Cancer Center Research Program, Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD 20817, USA; Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., Bethesda, MD 20817, USACenter for Prostate Disease Research, Murtha Cancer Center Research Program, Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD 20817, USA; Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., Bethesda, MD 20817, USACenter for Prostate Disease Research, Murtha Cancer Center Research Program, Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD 20817, USA; Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., Bethesda, MD 20817, USACenter for Prostate Disease Research, Murtha Cancer Center Research Program, Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD 20817, USA; Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., Bethesda, MD 20817, USACenter for Prostate Disease Research, Murtha Cancer Center Research Program, Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD 20817, USA; Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., Bethesda, MD 20817, USACenter for Prostate Disease Research, Murtha Cancer Center Research Program, Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD 20817, USA; Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., Bethesda, MD 20817, USAThe Joint Pathology Center, Silver Spring, MD 20910, USAUrologic Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USACenter for Prostate Disease Research, Murtha Cancer Center Research Program, Department of Surgery, Uniformed Services University of the Health Sciences, Bethesda, MD 20817, USA; F. Edward Hebert School of Medicine, Uniformed Services University of the Health Sciences, Bethesda, MD 20814, USA; Urology Service, Walter Reed National Military Medical Center, Bethesda, MD 20814, USAArtificial Intelligence Resource, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USAArtificial Intelligence Resource, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USAThe Joint Pathology Center, Silver Spring, MD 20910, USAThe Gleason score is an important predictor of prognosis in prostate cancer. However, its subjective nature can result in over- or under-grading. Our objective was to train an artificial intelligence (AI)-based algorithm to grade prostate cancer in specimens from patients who underwent radical prostatectomy (RP) and to assess the correlation of AI-estimated proportions of different Gleason patterns with biochemical recurrence-free survival (RFS), metastasis-free survival (MFS), and overall survival (OS). Training and validation of algorithms for cancer detection and grading were completed with three large datasets containing a total of 580 whole-mount prostate slides from 191 RP patients at two centers and 6218 annotated needle biopsy slides from the publicly available Prostate Cancer Grading Assessment dataset. A cancer detection model was trained using MobileNetV3 on 0.5 mm × 0.5 mm cancer areas (tiles) captured at 10× magnification. For cancer grading, a Gleason pattern detector was trained on tiles using a ResNet50 convolutional neural network and a selective CutMix training strategy involving a mixture of real and artificial examples. This strategy resulted in improved model generalizability in the test set compared with three different control experiments when evaluated on both needle biopsy slides and whole-mount prostate slides from different centers. In an additional test cohort of RP patients who were clinically followed over 30 years, quantitative Gleason pattern AI estimates achieved concordance indexes of 0.69, 0.72, and 0.64 for predicting RFS, MFS, and OS times, outperforming the control experiments and International Society of Urological Pathology system (ISUP) grading by pathologists. Finally, unsupervised clustering of test RP patient specimens into low-, medium-, and high-risk groups based on AI-estimated proportions of each Gleason pattern resulted in significantly improved RFS and MFS stratification compared with ISUP grading. In summary, deep learning-based quantitative Gleason scoring using a selective CutMix training strategy may improve prognostication after prostate cancer surgery.http://www.sciencedirect.com/science/article/pii/S2153353924000208Prostate cancerDigital pathologyGleason gradingArtificial intelligence
spellingShingle Sushant Patkar
Stephanie Harmon
Isabell Sesterhenn
Rosina Lis
Maria Merino
Denise Young
G. Thomas Brown
Kimberly M. Greenfield
John D. McGeeney
Sally Elsamanoudi
Shyh-Han Tan
Cara Schafer
Jiji Jiang
Gyorgy Petrovics
Albert Dobi
Francisco J. Rentas
Peter A. Pinto
Gregory T. Chesnut
Peter Choyke
Baris Turkbey
Joel T. Moncur
A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer
Journal of Pathology Informatics
Prostate cancer
Digital pathology
Gleason grading
Artificial intelligence
title A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer
title_full A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer
title_fullStr A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer
title_full_unstemmed A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer
title_short A selective CutMix approach improves generalizability of deep learning-based grading and risk assessment of prostate cancer
title_sort selective cutmix approach improves generalizability of deep learning based grading and risk assessment of prostate cancer
topic Prostate cancer
Digital pathology
Gleason grading
Artificial intelligence
url http://www.sciencedirect.com/science/article/pii/S2153353924000208
work_keys_str_mv AT sushantpatkar aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT stephanieharmon aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT isabellsesterhenn aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT rosinalis aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT mariamerino aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT deniseyoung aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT gthomasbrown aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT kimberlymgreenfield aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT johndmcgeeney aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT sallyelsamanoudi aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT shyhhantan aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT caraschafer aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT jijijiang aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT gyorgypetrovics aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT albertdobi aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT franciscojrentas aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT peterapinto aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT gregorytchesnut aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT peterchoyke aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT baristurkbey aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT joeltmoncur aselectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT sushantpatkar selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT stephanieharmon selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT isabellsesterhenn selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT rosinalis selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT mariamerino selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT deniseyoung selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT gthomasbrown selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT kimberlymgreenfield selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT johndmcgeeney selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT sallyelsamanoudi selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT shyhhantan selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT caraschafer selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT jijijiang selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT gyorgypetrovics selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT albertdobi selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT franciscojrentas selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT peterapinto selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT gregorytchesnut selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT peterchoyke selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT baristurkbey selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer
AT joeltmoncur selectivecutmixapproachimprovesgeneralizabilityofdeeplearningbasedgradingandriskassessmentofprostatecancer