Liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adults
Background: Traditional liver fibrosis staging via percutaneous biopsy suffers from sampling bias and variable inter-pathologist agreement, highlighting the need for more objective techniques. Deep learning models for disease staging from medical images have shown potential to decrease diagnostic va...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Journal of Pathology Informatics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2153353924000555 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850183007896338432 |
|---|---|
| author | Mahdieh Shabanian Zachary Taylor Christopher Woods Anas Bernieh Jonathan Dillman Lili He Sarangarajan Ranganathan Jennifer Picarsic Elanchezhian Somasundaram |
| author_facet | Mahdieh Shabanian Zachary Taylor Christopher Woods Anas Bernieh Jonathan Dillman Lili He Sarangarajan Ranganathan Jennifer Picarsic Elanchezhian Somasundaram |
| author_sort | Mahdieh Shabanian |
| collection | DOAJ |
| description | Background: Traditional liver fibrosis staging via percutaneous biopsy suffers from sampling bias and variable inter-pathologist agreement, highlighting the need for more objective techniques. Deep learning models for disease staging from medical images have shown potential to decrease diagnostic variability, with recent weakly supervised learning strategies showing promising results even with limited manual annotation. Purpose: To study the clustering-constrained attention multiple instance learning (CLAM) approach for staging liver fibrosis on trichrome whole slide images (WSIs) of children and young adults. Methods: This is an ethics board approved retrospective study utilizing 217 trichrome WSI from pediatric liver biopsies for model development and testing. Two pediatric pathologists scored WSI using two liver fibrosis staging systems, METAVIR and Ishak. Cases were then secondarily categorized into either high- or low-stage liver fibrosis and used for model development. The CLAM pipeline was used to develop binary classification models for histological liver fibrosis. Model performance was evaluated using area under the curve (AUC), accuracy, sensitivity, specificity, and Cohen's Kappa. Results: The CLAM models showed strong diagnostic performance, with sensitivities up to 0.76 and AUCs up to 0.92 for distinguishing low- and high-stage fibrosis. The agreement between model predictions and average pathologist scores was moderate to substantial (Kappa: 0.57–0.69), whereas pathologist agreement on the METAVIR and Ishak scoring systems was only fair (Kappa: 0.39–0.46). Conclusions: CLAM pipeline showed promise in detecting features important for differentiating low- and high-stage fibrosis from trichrome WSI based on the results, offering a promising objective method for liver fibrosis detection in children and young adults. |
| format | Article |
| id | doaj-art-641c7e15c0f34ece822adcdfa890a2bb |
| institution | OA Journals |
| issn | 2153-3539 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Pathology Informatics |
| spelling | doaj-art-641c7e15c0f34ece822adcdfa890a2bb2025-08-20T02:17:28ZengElsevierJournal of Pathology Informatics2153-35392025-01-011610041610.1016/j.jpi.2024.100416Liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adultsMahdieh Shabanian0Zachary Taylor1Christopher Woods2Anas Bernieh3Jonathan Dillman4Lili He5Sarangarajan Ranganathan6Jennifer Picarsic7Elanchezhian Somasundaram8University of Utah, Biomedical Informatics Department, Salt Lake City, UT, United StatesCincinnati Children's AI Imaging Research (CAIIR) Center, Cincinnati, OH, United StatesCincinnati Children's AI Imaging Research (CAIIR) Center, Cincinnati, OH, United States; Cincinnati Children's Hospital Division of Pathology, University of Cincinnati College of Medicine, Cincinnati, OH, United StatesCincinnati Children's Hospital Division of Pathology, University of Cincinnati College of Medicine, Cincinnati, OH, United StatesCincinnati Children's AI Imaging Research (CAIIR) Center, Cincinnati, OH, United States; Department of Radiology, Cincinnati Children's Hospital Medical Center, University of Cincinnati College of Medicine, Cincinnati, OH, United StatesCincinnati Children's AI Imaging Research (CAIIR) Center, Cincinnati, OH, United States; Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, United StatesCincinnati Children's Hospital Division of Pathology, University of Cincinnati College of Medicine, Cincinnati, OH, United States; Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, United StatesCincinnati Children's AI Imaging Research (CAIIR) Center, Cincinnati, OH, United States; Cincinnati Children's Hospital Division of Pathology, University of Cincinnati College of Medicine, Cincinnati, OH, United States; Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, United States; Department of Pathology, University of Pittsburgh School of Medicine, UPMC Children's Hospital, Pittsburgh, PA, United StatesCincinnati Children's AI Imaging Research (CAIIR) Center, Cincinnati, OH, United States; Department of Radiology, Cincinnati Children's Hospital Medical Center, University of Cincinnati College of Medicine, Cincinnati, OH, United States; Corresponding author at: Cincinnati Children's Hospital Medical Center, Radiology Department, 3333 Burnet Avenue, MLC 5033, Cincinnati, OH 45229, United States.Background: Traditional liver fibrosis staging via percutaneous biopsy suffers from sampling bias and variable inter-pathologist agreement, highlighting the need for more objective techniques. Deep learning models for disease staging from medical images have shown potential to decrease diagnostic variability, with recent weakly supervised learning strategies showing promising results even with limited manual annotation. Purpose: To study the clustering-constrained attention multiple instance learning (CLAM) approach for staging liver fibrosis on trichrome whole slide images (WSIs) of children and young adults. Methods: This is an ethics board approved retrospective study utilizing 217 trichrome WSI from pediatric liver biopsies for model development and testing. Two pediatric pathologists scored WSI using two liver fibrosis staging systems, METAVIR and Ishak. Cases were then secondarily categorized into either high- or low-stage liver fibrosis and used for model development. The CLAM pipeline was used to develop binary classification models for histological liver fibrosis. Model performance was evaluated using area under the curve (AUC), accuracy, sensitivity, specificity, and Cohen's Kappa. Results: The CLAM models showed strong diagnostic performance, with sensitivities up to 0.76 and AUCs up to 0.92 for distinguishing low- and high-stage fibrosis. The agreement between model predictions and average pathologist scores was moderate to substantial (Kappa: 0.57–0.69), whereas pathologist agreement on the METAVIR and Ishak scoring systems was only fair (Kappa: 0.39–0.46). Conclusions: CLAM pipeline showed promise in detecting features important for differentiating low- and high-stage fibrosis from trichrome WSI based on the results, offering a promising objective method for liver fibrosis detection in children and young adults.http://www.sciencedirect.com/science/article/pii/S2153353924000555Deep learningLiver fibrosisTrichromePediatricsMETAVIRIshak |
| spellingShingle | Mahdieh Shabanian Zachary Taylor Christopher Woods Anas Bernieh Jonathan Dillman Lili He Sarangarajan Ranganathan Jennifer Picarsic Elanchezhian Somasundaram Liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adults Journal of Pathology Informatics Deep learning Liver fibrosis Trichrome Pediatrics METAVIR Ishak |
| title | Liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adults |
| title_full | Liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adults |
| title_fullStr | Liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adults |
| title_full_unstemmed | Liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adults |
| title_short | Liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adults |
| title_sort | liver fibrosis classification on trichrome histology slides using weakly supervised learning in children and young adults |
| topic | Deep learning Liver fibrosis Trichrome Pediatrics METAVIR Ishak |
| url | http://www.sciencedirect.com/science/article/pii/S2153353924000555 |
| work_keys_str_mv | AT mahdiehshabanian liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults AT zacharytaylor liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults AT christopherwoods liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults AT anasbernieh liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults AT jonathandillman liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults AT lilihe liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults AT sarangarajanranganathan liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults AT jenniferpicarsic liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults AT elanchezhiansomasundaram liverfibrosisclassificationontrichromehistologyslidesusingweaklysupervisedlearninginchildrenandyoungadults |