Deep learning-based quantitative analysis of glomerular morphology in IgA nephropathy whole slide images and its prognostic implications
Abstract Kidney pathology of immunoglobulin A nephropathy (IgAN), which is the key finding of both diagnosis and risk stratification, involves labor-intensive manual interpretation as well as unavoidable interpreter-dependent variabilities. We propose artificial intelligence-based frameworks for qua...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-025-09031-w |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849238397229465600 |
|---|---|
| author | Seung Yeon Cho Yisak Kim Sehoon Park Jin Ho Paik Ho Jun Chin Jeong Hwan Park Jung Pyo Lee Yong-Jin Kim Sun-Hee Park Ho-chang Lee Hyunjeong Cho Beom Jin Lim Hyung Woo Kim Seung Hyeok Han Heounjeong Go Chung Hee Baek Hajeong Lee Kyung Chul Moon Young-Gon Kim |
| author_facet | Seung Yeon Cho Yisak Kim Sehoon Park Jin Ho Paik Ho Jun Chin Jeong Hwan Park Jung Pyo Lee Yong-Jin Kim Sun-Hee Park Ho-chang Lee Hyunjeong Cho Beom Jin Lim Hyung Woo Kim Seung Hyeok Han Heounjeong Go Chung Hee Baek Hajeong Lee Kyung Chul Moon Young-Gon Kim |
| author_sort | Seung Yeon Cho |
| collection | DOAJ |
| description | Abstract Kidney pathology of immunoglobulin A nephropathy (IgAN), which is the key finding of both diagnosis and risk stratification, involves labor-intensive manual interpretation as well as unavoidable interpreter-dependent variabilities. We propose artificial intelligence-based frameworks for quantitatively analyzing glomerular histologic features that can predict kidney progression in IgAN. A deep learning model, based on DeepLabV3Plus and EfficientNet-B3, was developed for segmenting glomeruli and quantifying the morphological features by using digitized whole slide images from seven tertiary hospitals. Subsequently, it was used for machine learning-based risk prediction of IgAN progression. Its predictability was compared with the conventional clinicopathologic feature-based model to demonstrate its comparable performance. In total, 1,241 whole slide images were obtained. The weighted averages of average precision and dice similarity coefficient were 0.795 and 0.721 in internal validation and 0.818 and 0.743 in external validation, respectively. Interestingly, image features-only-based kidney outcome prediction models showed similar predictability compared with clinical features-only-based models. In addition, incorporating an image-based deep learning model into the clinical features-based models enhanced predictabilities, although insignificant. These results show that quantitative glomerular histologic features are comparable to clinical data, suggesting that they may offer additional prognostic insights not covered by Oxford classification or other clinical parameters. |
| format | Article |
| id | doaj-art-a87fa6075be1411c92d97fcf9b7e90b0 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-a87fa6075be1411c92d97fcf9b7e90b02025-08-20T04:01:36ZengNature PortfolioScientific Reports2045-23222025-07-0115111210.1038/s41598-025-09031-wDeep learning-based quantitative analysis of glomerular morphology in IgA nephropathy whole slide images and its prognostic implicationsSeung Yeon Cho0Yisak Kim1Sehoon Park2Jin Ho Paik3Ho Jun Chin4Jeong Hwan Park5Jung Pyo Lee6Yong-Jin Kim7Sun-Hee Park8Ho-chang Lee9Hyunjeong Cho10Beom Jin Lim11Hyung Woo Kim12Seung Hyeok Han13Heounjeong Go14Chung Hee Baek15Hajeong Lee16Kyung Chul Moon17Young-Gon Kim18Interdisciplinary Program in Bioengineering, Graduate School, Seoul National UniversityInterdisciplinary Program in Bioengineering, Graduate School, Seoul National UniversityDepartment of Internal Medicine, Seoul National University HospitalDepartment of Pathology, Seoul National University Bundang Hospital, Seoul National University College of MedicineDepartment of Internal Medicine, Seoul National University College of MedicineDepartment of Pathology, Seoul Metropolitan Government Seoul National University Boramae Medical CenterDepartment of Internal Medicine, Seoul National University Boramae Medical CenterDepartment of Pathology, School of Medicine, Kyungpook National University Hospital, Kyungpook National UniversityDepartment of Internal Medicine, Kyungpook National University School of Medicine, Kyungpook National University HospitalDepartment of Pathology, Chungbuk National University College of MedicineDepartment of Internal Medicine, Chungbuk National University Hospital, Chungbuk National University College of MedicineDepartment of Pathology, Yonsei University College of MedicineDepartment of Internal Medicine, Institute of Kidney Disease Research, Yonsei University College of MedicineDepartment of Internal Medicine, Institute of Kidney Disease Research, Yonsei University College of MedicineDepartment of Pathology, Asan Medical Center, University of Ulsan College of MedicineDepartment of Internal Medicine, Asan Medical Center, University of Ulsan College of MedicineDepartment of Internal Medicine, Seoul National University HospitalDepartment of Pathology, Seoul National University College of MedicineDepartment of Transdisciplinary Medicine, Seoul National University HospitalAbstract Kidney pathology of immunoglobulin A nephropathy (IgAN), which is the key finding of both diagnosis and risk stratification, involves labor-intensive manual interpretation as well as unavoidable interpreter-dependent variabilities. We propose artificial intelligence-based frameworks for quantitatively analyzing glomerular histologic features that can predict kidney progression in IgAN. A deep learning model, based on DeepLabV3Plus and EfficientNet-B3, was developed for segmenting glomeruli and quantifying the morphological features by using digitized whole slide images from seven tertiary hospitals. Subsequently, it was used for machine learning-based risk prediction of IgAN progression. Its predictability was compared with the conventional clinicopathologic feature-based model to demonstrate its comparable performance. In total, 1,241 whole slide images were obtained. The weighted averages of average precision and dice similarity coefficient were 0.795 and 0.721 in internal validation and 0.818 and 0.743 in external validation, respectively. Interestingly, image features-only-based kidney outcome prediction models showed similar predictability compared with clinical features-only-based models. In addition, incorporating an image-based deep learning model into the clinical features-based models enhanced predictabilities, although insignificant. These results show that quantitative glomerular histologic features are comparable to clinical data, suggesting that they may offer additional prognostic insights not covered by Oxford classification or other clinical parameters.https://doi.org/10.1038/s41598-025-09031-w |
| spellingShingle | Seung Yeon Cho Yisak Kim Sehoon Park Jin Ho Paik Ho Jun Chin Jeong Hwan Park Jung Pyo Lee Yong-Jin Kim Sun-Hee Park Ho-chang Lee Hyunjeong Cho Beom Jin Lim Hyung Woo Kim Seung Hyeok Han Heounjeong Go Chung Hee Baek Hajeong Lee Kyung Chul Moon Young-Gon Kim Deep learning-based quantitative analysis of glomerular morphology in IgA nephropathy whole slide images and its prognostic implications Scientific Reports |
| title | Deep learning-based quantitative analysis of glomerular morphology in IgA nephropathy whole slide images and its prognostic implications |
| title_full | Deep learning-based quantitative analysis of glomerular morphology in IgA nephropathy whole slide images and its prognostic implications |
| title_fullStr | Deep learning-based quantitative analysis of glomerular morphology in IgA nephropathy whole slide images and its prognostic implications |
| title_full_unstemmed | Deep learning-based quantitative analysis of glomerular morphology in IgA nephropathy whole slide images and its prognostic implications |
| title_short | Deep learning-based quantitative analysis of glomerular morphology in IgA nephropathy whole slide images and its prognostic implications |
| title_sort | deep learning based quantitative analysis of glomerular morphology in iga nephropathy whole slide images and its prognostic implications |
| url | https://doi.org/10.1038/s41598-025-09031-w |
| work_keys_str_mv | AT seungyeoncho deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT yisakkim deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT sehoonpark deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT jinhopaik deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT hojunchin deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT jeonghwanpark deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT jungpyolee deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT yongjinkim deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT sunheepark deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT hochanglee deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT hyunjeongcho deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT beomjinlim deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT hyungwookim deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT seunghyeokhan deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT heounjeonggo deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT chungheebaek deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT hajeonglee deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT kyungchulmoon deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications AT younggonkim deeplearningbasedquantitativeanalysisofglomerularmorphologyiniganephropathywholeslideimagesanditsprognosticimplications |