Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions
Abstract Hematoxylin and eosin-stained biopsy slides are regularly available for colorectal cancer patients. These slides are often not used to define objective biomarkers for patient stratification and treatment selection. Standard biomarkers often pertain to costly and slow genetic tests. However,...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Portfolio
2023-09-01
|
Series: | Scientific Reports |
Online Access: | https://doi.org/10.1038/s41598-023-42357-x |
_version_ | 1797452937322037248 |
---|---|
author | Fabi Prezja Sami Äyrämö Ilkka Pölönen Timo Ojala Suvi Lahtinen Pekka Ruusuvuori Teijo Kuopio |
author_facet | Fabi Prezja Sami Äyrämö Ilkka Pölönen Timo Ojala Suvi Lahtinen Pekka Ruusuvuori Teijo Kuopio |
author_sort | Fabi Prezja |
collection | DOAJ |
description | Abstract Hematoxylin and eosin-stained biopsy slides are regularly available for colorectal cancer patients. These slides are often not used to define objective biomarkers for patient stratification and treatment selection. Standard biomarkers often pertain to costly and slow genetic tests. However, recent work has shown that relevant biomarkers can be extracted from these images using convolutional neural networks (CNNs). The CNN-based biomarkers predicted colorectal cancer patient outcomes comparably to gold standards. Extracting CNN-biomarkers is fast, automatic, and of minimal cost. CNN-based biomarkers rely on the ability of CNNs to recognize distinct tissue types from microscope whole slide images. The quality of these biomarkers (coined ‘Deep Stroma’) depends on the accuracy of CNNs in decomposing all relevant tissue classes. Improving tissue decomposition accuracy is essential for improving the prognostic potential of CNN-biomarkers. In this study, we implemented a novel training strategy to refine an established CNN model, which then surpassed all previous solutions . We obtained a 95.6% average accuracy in the external test set and 99.5% in the internal test set. Our approach reduced errors in biomarker-relevant classes, such as Lymphocytes, and was the first to include interpretability methods. These methods were used to better apprehend our model’s limitations and capabilities. |
first_indexed | 2024-03-09T15:15:54Z |
format | Article |
id | doaj.art-b9c09d0da1fc45be8e6e28e91ba6522f |
institution | Directory Open Access Journal |
issn | 2045-2322 |
language | English |
last_indexed | 2024-03-09T15:15:54Z |
publishDate | 2023-09-01 |
publisher | Nature Portfolio |
record_format | Article |
series | Scientific Reports |
spelling | doaj.art-b9c09d0da1fc45be8e6e28e91ba6522f2023-11-26T13:07:01ZengNature PortfolioScientific Reports2045-23222023-09-0113111410.1038/s41598-023-42357-xImproved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutionsFabi Prezja0Sami Äyrämö1Ilkka Pölönen2Timo Ojala3Suvi Lahtinen4Pekka Ruusuvuori5Teijo Kuopio6Faculty of Information Technology, University of JyväskyläFaculty of Information Technology, University of JyväskyläFaculty of Information Technology, University of JyväskyläFaculty of Information Technology, University of JyväskyläFaculty of Information Technology, University of JyväskyläInstitute of Biomedicine, Cancer Research Unit, University of TurkuDepartment of Education and Research, Hospital Nova of Central FinlandAbstract Hematoxylin and eosin-stained biopsy slides are regularly available for colorectal cancer patients. These slides are often not used to define objective biomarkers for patient stratification and treatment selection. Standard biomarkers often pertain to costly and slow genetic tests. However, recent work has shown that relevant biomarkers can be extracted from these images using convolutional neural networks (CNNs). The CNN-based biomarkers predicted colorectal cancer patient outcomes comparably to gold standards. Extracting CNN-biomarkers is fast, automatic, and of minimal cost. CNN-based biomarkers rely on the ability of CNNs to recognize distinct tissue types from microscope whole slide images. The quality of these biomarkers (coined ‘Deep Stroma’) depends on the accuracy of CNNs in decomposing all relevant tissue classes. Improving tissue decomposition accuracy is essential for improving the prognostic potential of CNN-biomarkers. In this study, we implemented a novel training strategy to refine an established CNN model, which then surpassed all previous solutions . We obtained a 95.6% average accuracy in the external test set and 99.5% in the internal test set. Our approach reduced errors in biomarker-relevant classes, such as Lymphocytes, and was the first to include interpretability methods. These methods were used to better apprehend our model’s limitations and capabilities.https://doi.org/10.1038/s41598-023-42357-x |
spellingShingle | Fabi Prezja Sami Äyrämö Ilkka Pölönen Timo Ojala Suvi Lahtinen Pekka Ruusuvuori Teijo Kuopio Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions Scientific Reports |
title | Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions |
title_full | Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions |
title_fullStr | Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions |
title_full_unstemmed | Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions |
title_short | Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions |
title_sort | improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions |
url | https://doi.org/10.1038/s41598-023-42357-x |
work_keys_str_mv | AT fabiprezja improvedaccuracyincolorectalcancertissuedecompositionthroughrefinementofestablisheddeeplearningsolutions AT samiayramo improvedaccuracyincolorectalcancertissuedecompositionthroughrefinementofestablisheddeeplearningsolutions AT ilkkapolonen improvedaccuracyincolorectalcancertissuedecompositionthroughrefinementofestablisheddeeplearningsolutions AT timoojala improvedaccuracyincolorectalcancertissuedecompositionthroughrefinementofestablisheddeeplearningsolutions AT suvilahtinen improvedaccuracyincolorectalcancertissuedecompositionthroughrefinementofestablisheddeeplearningsolutions AT pekkaruusuvuori improvedaccuracyincolorectalcancertissuedecompositionthroughrefinementofestablisheddeeplearningsolutions AT teijokuopio improvedaccuracyincolorectalcancertissuedecompositionthroughrefinementofestablisheddeeplearningsolutions |