Mapping cell-to-tissue graphs across human placenta histology whole slide images using deep learning with HAPPY
Abstract Accurate placenta pathology assessment is essential for managing maternal and newborn health, but the placenta’s heterogeneity and temporal variability pose challenges for histology analysis. To address this issue, we developed the ‘Histology Analysis Pipeline.PY’ (HAPPY), a deep learning h...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-03-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-46986-2 |
| _version_ | 1797233494104997888 |
|---|---|
| author | Claudia Vanea Jelisaveta Džigurski Valentina Rukins Omri Dodi Siim Siigur Liis Salumäe Karen Meir W. Tony Parks Drorith Hochner-Celnikier Abigail Fraser Hagit Hochner Triin Laisk Linda M. Ernst Cecilia M. Lindgren Christoffer Nellåker |
| author_facet | Claudia Vanea Jelisaveta Džigurski Valentina Rukins Omri Dodi Siim Siigur Liis Salumäe Karen Meir W. Tony Parks Drorith Hochner-Celnikier Abigail Fraser Hagit Hochner Triin Laisk Linda M. Ernst Cecilia M. Lindgren Christoffer Nellåker |
| author_sort | Claudia Vanea |
| collection | DOAJ |
| description | Abstract Accurate placenta pathology assessment is essential for managing maternal and newborn health, but the placenta’s heterogeneity and temporal variability pose challenges for histology analysis. To address this issue, we developed the ‘Histology Analysis Pipeline.PY’ (HAPPY), a deep learning hierarchical method for quantifying the variability of cells and micro-anatomical tissue structures across placenta histology whole slide images. HAPPY differs from patch-based features or segmentation approaches by following an interpretable biological hierarchy, representing cells and cellular communities within tissues at a single-cell resolution across whole slide images. We present a set of quantitative metrics from healthy term placentas as a baseline for future assessments of placenta health and we show how these metrics deviate in placentas with clinically significant placental infarction. HAPPY’s cell and tissue predictions closely replicate those from independent clinical experts and placental biology literature. |
| first_indexed | 2024-04-24T16:17:03Z |
| format | Article |
| id | doaj.art-dd7e5a53a31c4a3c9e8c1a2077757cd7 |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-04-24T16:17:03Z |
| publishDate | 2024-03-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-dd7e5a53a31c4a3c9e8c1a2077757cd72024-03-31T11:25:44ZengNature PortfolioNature Communications2041-17232024-03-0115111610.1038/s41467-024-46986-2Mapping cell-to-tissue graphs across human placenta histology whole slide images using deep learning with HAPPYClaudia Vanea0Jelisaveta Džigurski1Valentina Rukins2Omri Dodi3Siim Siigur4Liis Salumäe5Karen Meir6W. Tony Parks7Drorith Hochner-Celnikier8Abigail Fraser9Hagit Hochner10Triin Laisk11Linda M. Ernst12Cecilia M. Lindgren13Christoffer Nellåker14Nuffield Department of Women’s & Reproductive Health, University of OxfordInstitute of Genomics, University of TartuInstitute of Genomics, University of TartuFaculty of Medicine, Hadassah Hebrew University Medical CenterDepartment of Pathology, Tartu University HospitalDepartment of Pathology, Tartu University HospitalDepartment of Pathology, Hadassah Hebrew University Medical CenterDepartment of Laboratory Medicine & Pathobiology, University of TorontoFaculty of Medicine, Hadassah Hebrew University Medical CenterPopulation Health Sciences, Bristol Medical School, University of BristolBraun School of Public Health, Hebrew University of JerusalemInstitute of Genomics, University of TartuDepartment of Pathology and Laboratory Medicine, NorthShore University HealthSystemBig Data Institute, Li Ka Shing Centre for Health Information and Discovery, University of OxfordNuffield Department of Women’s & Reproductive Health, University of OxfordAbstract Accurate placenta pathology assessment is essential for managing maternal and newborn health, but the placenta’s heterogeneity and temporal variability pose challenges for histology analysis. To address this issue, we developed the ‘Histology Analysis Pipeline.PY’ (HAPPY), a deep learning hierarchical method for quantifying the variability of cells and micro-anatomical tissue structures across placenta histology whole slide images. HAPPY differs from patch-based features or segmentation approaches by following an interpretable biological hierarchy, representing cells and cellular communities within tissues at a single-cell resolution across whole slide images. We present a set of quantitative metrics from healthy term placentas as a baseline for future assessments of placenta health and we show how these metrics deviate in placentas with clinically significant placental infarction. HAPPY’s cell and tissue predictions closely replicate those from independent clinical experts and placental biology literature.https://doi.org/10.1038/s41467-024-46986-2 |
| spellingShingle | Claudia Vanea Jelisaveta Džigurski Valentina Rukins Omri Dodi Siim Siigur Liis Salumäe Karen Meir W. Tony Parks Drorith Hochner-Celnikier Abigail Fraser Hagit Hochner Triin Laisk Linda M. Ernst Cecilia M. Lindgren Christoffer Nellåker Mapping cell-to-tissue graphs across human placenta histology whole slide images using deep learning with HAPPY Nature Communications |
| title | Mapping cell-to-tissue graphs across human placenta histology whole slide images using deep learning with HAPPY |
| title_full | Mapping cell-to-tissue graphs across human placenta histology whole slide images using deep learning with HAPPY |
| title_fullStr | Mapping cell-to-tissue graphs across human placenta histology whole slide images using deep learning with HAPPY |
| title_full_unstemmed | Mapping cell-to-tissue graphs across human placenta histology whole slide images using deep learning with HAPPY |
| title_short | Mapping cell-to-tissue graphs across human placenta histology whole slide images using deep learning with HAPPY |
| title_sort | mapping cell to tissue graphs across human placenta histology whole slide images using deep learning with happy |
| url | https://doi.org/10.1038/s41467-024-46986-2 |
| work_keys_str_mv | AT claudiavanea mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT jelisavetadzigurski mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT valentinarukins mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT omridodi mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT siimsiigur mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT liissalumae mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT karenmeir mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT wtonyparks mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT drorithhochnercelnikier mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT abigailfraser mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT hagithochner mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT triinlaisk mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT lindamernst mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT ceciliamlindgren mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy AT christoffernellaker mappingcelltotissuegraphsacrosshumanplacentahistologywholeslideimagesusingdeeplearningwithhappy |