Impact of stain variation and color normalization for prognostic predictions in pathology
Abstract In recent years, deep neural networks (DNNs) have demonstrated remarkable performance in pathology applications, potentially even outperforming expert pathologists due to their ability to learn subtle features from large datasets. One complication in preparing digital pathology datasets for...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-024-83267-w |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594737388847104 |
|---|---|
| author | Siyu Lin Haowen Zhou Mark Watson Ramaswamy Govindan Richard J. Cote Changhuei Yang |
| author_facet | Siyu Lin Haowen Zhou Mark Watson Ramaswamy Govindan Richard J. Cote Changhuei Yang |
| author_sort | Siyu Lin |
| collection | DOAJ |
| description | Abstract In recent years, deep neural networks (DNNs) have demonstrated remarkable performance in pathology applications, potentially even outperforming expert pathologists due to their ability to learn subtle features from large datasets. One complication in preparing digital pathology datasets for DNN tasks is the variation in tinctorial qualities. A common way to address this is to perform stain normalization on the images. In this study, we show that a well-trained DNN model trained on one batch of histological slides failed to generalize to another batch prepared at a different time from the same tissue blocks, even when stain normalization methods were applied. This study used sample data from a previously reported DNN that was able to identify patients with early-stage non-small cell lung cancer (NSCLC) whose tumors did and did not metastasize, with high accuracy, based on training and then testing of digital images from H&E stained primary tumor tissue sections processed at the same time. In this study, we obtained a new series of histologic slides from the adjacent recuts of the same tissue blocks processed in the same lab but at a different time. We found that the DNN trained on either batch of slides/images was unable to generalize and failed to predict progression in the other batch of slides/images (AUCcross-batch = 0.52 - 0.53 compared to AUCsame-batch = 0.74 - 0.81). The failure to generalize did not improve even when the tinctorial difference corrections were made through either traditional color-tuning or stain normalization with the help of a Cycle Generative Adversarial Network (CycleGAN) process. This highlights the need to develop an entirely new way to process and collect consistent microscopy images from histologic slides that can be used to both train and allow for the general application of predictive DNN algorithms. |
| format | Article |
| id | doaj-art-0532d7b9d5f64cf1a81fb4f78fdcceb8 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-0532d7b9d5f64cf1a81fb4f78fdcceb82025-01-19T12:22:11ZengNature PortfolioScientific Reports2045-23222025-01-0115111010.1038/s41598-024-83267-wImpact of stain variation and color normalization for prognostic predictions in pathologySiyu Lin0Haowen Zhou1Mark Watson2Ramaswamy Govindan3Richard J. Cote4Changhuei Yang5Department of Electrical Engineering, California Institute of TechnologyDepartment of Electrical Engineering, California Institute of TechnologyDepartment of Pathology and Immunology, Washington University School of MedicineDepartment of Medicine, Washington University School of MedicineDepartment of Pathology and Immunology, Washington University School of MedicineDepartment of Electrical Engineering, California Institute of TechnologyAbstract In recent years, deep neural networks (DNNs) have demonstrated remarkable performance in pathology applications, potentially even outperforming expert pathologists due to their ability to learn subtle features from large datasets. One complication in preparing digital pathology datasets for DNN tasks is the variation in tinctorial qualities. A common way to address this is to perform stain normalization on the images. In this study, we show that a well-trained DNN model trained on one batch of histological slides failed to generalize to another batch prepared at a different time from the same tissue blocks, even when stain normalization methods were applied. This study used sample data from a previously reported DNN that was able to identify patients with early-stage non-small cell lung cancer (NSCLC) whose tumors did and did not metastasize, with high accuracy, based on training and then testing of digital images from H&E stained primary tumor tissue sections processed at the same time. In this study, we obtained a new series of histologic slides from the adjacent recuts of the same tissue blocks processed in the same lab but at a different time. We found that the DNN trained on either batch of slides/images was unable to generalize and failed to predict progression in the other batch of slides/images (AUCcross-batch = 0.52 - 0.53 compared to AUCsame-batch = 0.74 - 0.81). The failure to generalize did not improve even when the tinctorial difference corrections were made through either traditional color-tuning or stain normalization with the help of a Cycle Generative Adversarial Network (CycleGAN) process. This highlights the need to develop an entirely new way to process and collect consistent microscopy images from histologic slides that can be used to both train and allow for the general application of predictive DNN algorithms.https://doi.org/10.1038/s41598-024-83267-w |
| spellingShingle | Siyu Lin Haowen Zhou Mark Watson Ramaswamy Govindan Richard J. Cote Changhuei Yang Impact of stain variation and color normalization for prognostic predictions in pathology Scientific Reports |
| title | Impact of stain variation and color normalization for prognostic predictions in pathology |
| title_full | Impact of stain variation and color normalization for prognostic predictions in pathology |
| title_fullStr | Impact of stain variation and color normalization for prognostic predictions in pathology |
| title_full_unstemmed | Impact of stain variation and color normalization for prognostic predictions in pathology |
| title_short | Impact of stain variation and color normalization for prognostic predictions in pathology |
| title_sort | impact of stain variation and color normalization for prognostic predictions in pathology |
| url | https://doi.org/10.1038/s41598-024-83267-w |
| work_keys_str_mv | AT siyulin impactofstainvariationandcolornormalizationforprognosticpredictionsinpathology AT haowenzhou impactofstainvariationandcolornormalizationforprognosticpredictionsinpathology AT markwatson impactofstainvariationandcolornormalizationforprognosticpredictionsinpathology AT ramaswamygovindan impactofstainvariationandcolornormalizationforprognosticpredictionsinpathology AT richardjcote impactofstainvariationandcolornormalizationforprognosticpredictionsinpathology AT changhueiyang impactofstainvariationandcolornormalizationforprognosticpredictionsinpathology |