Self-Organizing Maps for Cellular In Silico Staining and Cell Substate Classification
Cellular composition and structural organization of cells in the tissue determine effective antitumor response and can predict patient outcome and therapy response. Here we present Seg-SOM, a method for dimensionality reduction of cell morphology in H&E-stained tissue images. Seg-SOM resolve...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-10-01
|
| Series: | Frontiers in Immunology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fimmu.2021.765923/full |
| _version_ | 1818835843901554688 |
|---|---|
| author | Edwin Yuan Magdalena Matusiak Korsuk Sirinukunwattana Korsuk Sirinukunwattana Korsuk Sirinukunwattana Korsuk Sirinukunwattana Sushama Varma Łukasz Kidziński Robert West |
| author_facet | Edwin Yuan Magdalena Matusiak Korsuk Sirinukunwattana Korsuk Sirinukunwattana Korsuk Sirinukunwattana Korsuk Sirinukunwattana Sushama Varma Łukasz Kidziński Robert West |
| author_sort | Edwin Yuan |
| collection | DOAJ |
| description | Cellular composition and structural organization of cells in the tissue determine effective antitumor response and can predict patient outcome and therapy response. Here we present Seg-SOM, a method for dimensionality reduction of cell morphology in H&E-stained tissue images. Seg-SOM resolves cellular tissue heterogeneity and reveals complex tissue architecture. We leverage a self-organizing map (SOM) artificial neural network to group cells based on morphological features like shape and size. Seg-SOM allows for cell segmentation, systematic classification, and in silico cell labeling. We apply the Seg-SOM to a dataset of breast cancer progression images and find that clustering of SOM classes reveals groups of cells corresponding to fibroblasts, epithelial cells, and lymphocytes. We show that labeling the Lymphocyte SOM class on the breast tissue images accurately estimates lymphocytic infiltration. We further demonstrate how to use Seq-SOM in combination with non-negative matrix factorization to statistically describe the interaction of cell subtypes and use the interaction information as highly interpretable features for a histological classifier. Our work provides a framework for use of SOM in human pathology to resolve cellular composition of complex human tissues. We provide a python implementation and an easy-to-use docker deployment, enabling researchers to effortlessly featurize digitalized H&E-stained tissue. |
| first_indexed | 2024-12-19T02:57:10Z |
| format | Article |
| id | doaj.art-c9c959eabd42499e9f1f8a569043f3da |
| institution | Directory Open Access Journal |
| issn | 1664-3224 |
| language | English |
| last_indexed | 2024-12-19T02:57:10Z |
| publishDate | 2021-10-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Immunology |
| spelling | doaj.art-c9c959eabd42499e9f1f8a569043f3da2022-12-21T20:38:19ZengFrontiers Media S.A.Frontiers in Immunology1664-32242021-10-011210.3389/fimmu.2021.765923765923Self-Organizing Maps for Cellular In Silico Staining and Cell Substate ClassificationEdwin Yuan0Magdalena Matusiak1Korsuk Sirinukunwattana2Korsuk Sirinukunwattana3Korsuk Sirinukunwattana4Korsuk Sirinukunwattana5Sushama Varma6Łukasz Kidziński7Robert West8Department of Applied Physics, Stanford University, Stanford, CA, United StatesDepartment of Pathology, Stanford University, Stanford, CA, United StatesInstitute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, United KingdomGround Truth Labs, Oxford, United KingdomBig Data Institute/Li Ka Shing Centre for Health Information and Discovery, University of Oxford, Oxford, United KingdomNational Institute for Health Research (NIHR) Oxford Biomedical Research Centre, Oxford University National Health Service (NHS) Foundation Trust, Oxford, United KingdomDepartment of Pathology, Stanford University, Stanford, CA, United StatesDepartment of Bioengineering, Stanford University, Stanford, CA, United StatesDepartment of Pathology, Stanford University, Stanford, CA, United StatesCellular composition and structural organization of cells in the tissue determine effective antitumor response and can predict patient outcome and therapy response. Here we present Seg-SOM, a method for dimensionality reduction of cell morphology in H&E-stained tissue images. Seg-SOM resolves cellular tissue heterogeneity and reveals complex tissue architecture. We leverage a self-organizing map (SOM) artificial neural network to group cells based on morphological features like shape and size. Seg-SOM allows for cell segmentation, systematic classification, and in silico cell labeling. We apply the Seg-SOM to a dataset of breast cancer progression images and find that clustering of SOM classes reveals groups of cells corresponding to fibroblasts, epithelial cells, and lymphocytes. We show that labeling the Lymphocyte SOM class on the breast tissue images accurately estimates lymphocytic infiltration. We further demonstrate how to use Seq-SOM in combination with non-negative matrix factorization to statistically describe the interaction of cell subtypes and use the interaction information as highly interpretable features for a histological classifier. Our work provides a framework for use of SOM in human pathology to resolve cellular composition of complex human tissues. We provide a python implementation and an easy-to-use docker deployment, enabling researchers to effortlessly featurize digitalized H&E-stained tissue.https://www.frontiersin.org/articles/10.3389/fimmu.2021.765923/fullself-organizationin silico staininge-pathologycell subtype classificationsegmentation |
| spellingShingle | Edwin Yuan Magdalena Matusiak Korsuk Sirinukunwattana Korsuk Sirinukunwattana Korsuk Sirinukunwattana Korsuk Sirinukunwattana Sushama Varma Łukasz Kidziński Robert West Self-Organizing Maps for Cellular In Silico Staining and Cell Substate Classification Frontiers in Immunology self-organization in silico staining e-pathology cell subtype classification segmentation |
| title | Self-Organizing Maps for Cellular In Silico Staining and Cell Substate Classification |
| title_full | Self-Organizing Maps for Cellular In Silico Staining and Cell Substate Classification |
| title_fullStr | Self-Organizing Maps for Cellular In Silico Staining and Cell Substate Classification |
| title_full_unstemmed | Self-Organizing Maps for Cellular In Silico Staining and Cell Substate Classification |
| title_short | Self-Organizing Maps for Cellular In Silico Staining and Cell Substate Classification |
| title_sort | self organizing maps for cellular in silico staining and cell substate classification |
| topic | self-organization in silico staining e-pathology cell subtype classification segmentation |
| url | https://www.frontiersin.org/articles/10.3389/fimmu.2021.765923/full |
| work_keys_str_mv | AT edwinyuan selforganizingmapsforcellularinsilicostainingandcellsubstateclassification AT magdalenamatusiak selforganizingmapsforcellularinsilicostainingandcellsubstateclassification AT korsuksirinukunwattana selforganizingmapsforcellularinsilicostainingandcellsubstateclassification AT korsuksirinukunwattana selforganizingmapsforcellularinsilicostainingandcellsubstateclassification AT korsuksirinukunwattana selforganizingmapsforcellularinsilicostainingandcellsubstateclassification AT korsuksirinukunwattana selforganizingmapsforcellularinsilicostainingandcellsubstateclassification AT sushamavarma selforganizingmapsforcellularinsilicostainingandcellsubstateclassification AT łukaszkidzinski selforganizingmapsforcellularinsilicostainingandcellsubstateclassification AT robertwest selforganizingmapsforcellularinsilicostainingandcellsubstateclassification |