Deep learning models for histologic grading of breast cancer and association with disease prognosis
Abstract Histologic grading of breast cancer involves review and scoring of three well-established morphologic features: mitotic count, nuclear pleomorphism, and tubule formation. Taken together, these features form the basis of the Nottingham Grading System which is used to inform breast cancer cha...
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-10-01
|
| Series: | npj Breast Cancer |
| Online Access: | https://doi.org/10.1038/s41523-022-00478-y |
| _version_ | 1797641447580631040 |
|---|---|
| author | Ronnachai Jaroensri Ellery Wulczyn Narayan Hegde Trissia Brown Isabelle Flament-Auvigne Fraser Tan Yuannan Cai Kunal Nagpal Emad A. Rakha David J. Dabbs Niels Olson James H. Wren Elaine E. Thompson Erik Seetao Carrie Robinson Melissa Miao Fabien Beckers Greg S. Corrado Lily H. Peng Craig H. Mermel Yun Liu David F. Steiner Po-Hsuan Cameron Chen |
| author_facet | Ronnachai Jaroensri Ellery Wulczyn Narayan Hegde Trissia Brown Isabelle Flament-Auvigne Fraser Tan Yuannan Cai Kunal Nagpal Emad A. Rakha David J. Dabbs Niels Olson James H. Wren Elaine E. Thompson Erik Seetao Carrie Robinson Melissa Miao Fabien Beckers Greg S. Corrado Lily H. Peng Craig H. Mermel Yun Liu David F. Steiner Po-Hsuan Cameron Chen |
| author_sort | Ronnachai Jaroensri |
| collection | DOAJ |
| description | Abstract Histologic grading of breast cancer involves review and scoring of three well-established morphologic features: mitotic count, nuclear pleomorphism, and tubule formation. Taken together, these features form the basis of the Nottingham Grading System which is used to inform breast cancer characterization and prognosis. In this study, we develop deep learning models to perform histologic scoring of all three components using digitized hematoxylin and eosin-stained slides containing invasive breast carcinoma. We first evaluate model performance using pathologist-based reference standards for each component. To complement this typical approach to evaluation, we further evaluate the deep learning models via prognostic analyses. The individual component models perform at or above published benchmarks for algorithm-based grading approaches, achieving high concordance rates with pathologist grading. Further, prognostic performance using deep learning-based grading is on par with that of pathologists performing review of matched slides. By providing scores for each component feature, the deep-learning based approach also provides the potential to identify the grading components contributing most to prognostic value. This may enable optimized prognostic models, opportunities to improve access to consistent grading, and approaches to better understand the links between histologic features and clinical outcomes in breast cancer. |
| first_indexed | 2024-03-11T13:45:43Z |
| format | Article |
| id | doaj.art-e87e4501ec7945b0aef836b8987fc7e1 |
| institution | Directory Open Access Journal |
| issn | 2374-4677 |
| language | English |
| last_indexed | 2024-03-11T13:45:43Z |
| publishDate | 2022-10-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Breast Cancer |
| spelling | doaj.art-e87e4501ec7945b0aef836b8987fc7e12023-11-02T10:48:06ZengNature Portfolionpj Breast Cancer2374-46772022-10-018111210.1038/s41523-022-00478-yDeep learning models for histologic grading of breast cancer and association with disease prognosisRonnachai Jaroensri0Ellery Wulczyn1Narayan Hegde2Trissia Brown3Isabelle Flament-Auvigne4Fraser Tan5Yuannan Cai6Kunal Nagpal7Emad A. Rakha8David J. Dabbs9Niels Olson10James H. Wren11Elaine E. Thompson12Erik Seetao13Carrie Robinson14Melissa Miao15Fabien Beckers16Greg S. Corrado17Lily H. Peng18Craig H. Mermel19Yun Liu20David F. Steiner21Po-Hsuan Cameron Chen22Google HealthGoogle HealthGoogle HealthWork done at Google Health via VituityWork done at Google Health via VituityGoogle HealthGoogle HealthWork done at Google Health, current affiliation Tempus Labs IncDepartment of Pathology, School of Medicine, University of NottinghamJohn A. Burns University of Hawaii Cancer CenterDefense Innovation UnitHenry M. Jackson FoundationHenry M. Jackson FoundationHenry M. Jackson FoundationLaboratory Department, Naval Medical Center San DiegoVerily Life SciencesVerily Life SciencesGoogle HealthGoogle HealthGoogle HealthGoogle HealthGoogle HealthGoogle HealthAbstract Histologic grading of breast cancer involves review and scoring of three well-established morphologic features: mitotic count, nuclear pleomorphism, and tubule formation. Taken together, these features form the basis of the Nottingham Grading System which is used to inform breast cancer characterization and prognosis. In this study, we develop deep learning models to perform histologic scoring of all three components using digitized hematoxylin and eosin-stained slides containing invasive breast carcinoma. We first evaluate model performance using pathologist-based reference standards for each component. To complement this typical approach to evaluation, we further evaluate the deep learning models via prognostic analyses. The individual component models perform at or above published benchmarks for algorithm-based grading approaches, achieving high concordance rates with pathologist grading. Further, prognostic performance using deep learning-based grading is on par with that of pathologists performing review of matched slides. By providing scores for each component feature, the deep-learning based approach also provides the potential to identify the grading components contributing most to prognostic value. This may enable optimized prognostic models, opportunities to improve access to consistent grading, and approaches to better understand the links between histologic features and clinical outcomes in breast cancer.https://doi.org/10.1038/s41523-022-00478-y |
| spellingShingle | Ronnachai Jaroensri Ellery Wulczyn Narayan Hegde Trissia Brown Isabelle Flament-Auvigne Fraser Tan Yuannan Cai Kunal Nagpal Emad A. Rakha David J. Dabbs Niels Olson James H. Wren Elaine E. Thompson Erik Seetao Carrie Robinson Melissa Miao Fabien Beckers Greg S. Corrado Lily H. Peng Craig H. Mermel Yun Liu David F. Steiner Po-Hsuan Cameron Chen Deep learning models for histologic grading of breast cancer and association with disease prognosis npj Breast Cancer |
| title | Deep learning models for histologic grading of breast cancer and association with disease prognosis |
| title_full | Deep learning models for histologic grading of breast cancer and association with disease prognosis |
| title_fullStr | Deep learning models for histologic grading of breast cancer and association with disease prognosis |
| title_full_unstemmed | Deep learning models for histologic grading of breast cancer and association with disease prognosis |
| title_short | Deep learning models for histologic grading of breast cancer and association with disease prognosis |
| title_sort | deep learning models for histologic grading of breast cancer and association with disease prognosis |
| url | https://doi.org/10.1038/s41523-022-00478-y |
| work_keys_str_mv | AT ronnachaijaroensri deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT ellerywulczyn deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT narayanhegde deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT trissiabrown deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT isabelleflamentauvigne deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT frasertan deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT yuannancai deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT kunalnagpal deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT emadarakha deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT davidjdabbs deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT nielsolson deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT jameshwren deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT elaineethompson deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT erikseetao deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT carrierobinson deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT melissamiao deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT fabienbeckers deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT gregscorrado deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT lilyhpeng deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT craighmermel deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT yunliu deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT davidfsteiner deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis AT pohsuancameronchen deeplearningmodelsforhistologicgradingofbreastcancerandassociationwithdiseaseprognosis |