A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification
Abstract Histopathology foundation models show great promise across many tasks, but analyses have been limited by arbitrary hyperparameters. We report the most rigorous single-task validation study to date, specifically in the context of ovarian carcinoma morphological subtyping. Attention-based mul...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | npj Precision Oncology |
| Online Access: | https://doi.org/10.1038/s41698-025-00799-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832572064672776192 |
|---|---|
| author | Jack Breen Katie Allen Kieran Zucker Lucy Godson Nicolas M. Orsi Nishant Ravikumar |
| author_facet | Jack Breen Katie Allen Kieran Zucker Lucy Godson Nicolas M. Orsi Nishant Ravikumar |
| author_sort | Jack Breen |
| collection | DOAJ |
| description | Abstract Histopathology foundation models show great promise across many tasks, but analyses have been limited by arbitrary hyperparameters. We report the most rigorous single-task validation study to date, specifically in the context of ovarian carcinoma morphological subtyping. Attention-based multiple instance learning classifiers were compared using three ImageNet-pretrained encoders and fourteen foundation models, each trained with 1864 whole slide images and validated through hold-out testing and two external validations (the Transcanadian Study and OCEAN Challenge). The best-performing classifier used the H-optimus-0 foundation model, with balanced accuracies of 89%, 97%, and 74%, though UNI achieved similar results at a quarter of the computational cost. Hyperparameter tuning the classifiers improved performance by a median 1.9% balanced accuracy, with many improvements being statistically significant. Foundation models improve classification performance and may allow for clinical utility, with models providing a second opinion in challenging cases and potentially improving the accuracy and efficiency of diagnoses. |
| format | Article |
| id | doaj-art-462d352bde2749bb957cb0c0fd1d5590 |
| institution | Kabale University |
| issn | 2397-768X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Precision Oncology |
| spelling | doaj-art-462d352bde2749bb957cb0c0fd1d55902025-02-02T12:06:35ZengNature Portfolionpj Precision Oncology2397-768X2025-01-019111210.1038/s41698-025-00799-8A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classificationJack Breen0Katie Allen1Kieran Zucker2Lucy Godson3Nicolas M. Orsi4Nishant Ravikumar5Centre for Computational Imaging and Simulation Technologies in Biomedicine (CISTIB), School of Computing, University of LeedsLeeds Institute of Medical Research at St James’s, School of Medicine, University of LeedsLeeds Cancer Centre, St James’s University HospitalNational Pathology Imaging Cooperative (NPIC), Leeds Teaching Hospitals NHS TrustLeeds Institute of Medical Research at St James’s, School of Medicine, University of LeedsCentre for Computational Imaging and Simulation Technologies in Biomedicine (CISTIB), School of Computing, University of LeedsAbstract Histopathology foundation models show great promise across many tasks, but analyses have been limited by arbitrary hyperparameters. We report the most rigorous single-task validation study to date, specifically in the context of ovarian carcinoma morphological subtyping. Attention-based multiple instance learning classifiers were compared using three ImageNet-pretrained encoders and fourteen foundation models, each trained with 1864 whole slide images and validated through hold-out testing and two external validations (the Transcanadian Study and OCEAN Challenge). The best-performing classifier used the H-optimus-0 foundation model, with balanced accuracies of 89%, 97%, and 74%, though UNI achieved similar results at a quarter of the computational cost. Hyperparameter tuning the classifiers improved performance by a median 1.9% balanced accuracy, with many improvements being statistically significant. Foundation models improve classification performance and may allow for clinical utility, with models providing a second opinion in challenging cases and potentially improving the accuracy and efficiency of diagnoses.https://doi.org/10.1038/s41698-025-00799-8 |
| spellingShingle | Jack Breen Katie Allen Kieran Zucker Lucy Godson Nicolas M. Orsi Nishant Ravikumar A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification npj Precision Oncology |
| title | A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification |
| title_full | A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification |
| title_fullStr | A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification |
| title_full_unstemmed | A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification |
| title_short | A comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification |
| title_sort | comprehensive evaluation of histopathology foundation models for ovarian cancer subtype classification |
| url | https://doi.org/10.1038/s41698-025-00799-8 |
| work_keys_str_mv | AT jackbreen acomprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT katieallen acomprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT kieranzucker acomprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT lucygodson acomprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT nicolasmorsi acomprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT nishantravikumar acomprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT jackbreen comprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT katieallen comprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT kieranzucker comprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT lucygodson comprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT nicolasmorsi comprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification AT nishantravikumar comprehensiveevaluationofhistopathologyfoundationmodelsforovariancancersubtypeclassification |