Deep-learning model for evaluating histopathology of acute renal tubular injury
Abstract Tubular injury is the most common cause of acute kidney injury. Histopathological diagnosis may help distinguish between the different types of acute kidney injury and aid in treatment. To date, a limited number of study has used deep-learning models to assist in the histopathological diagn...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-04-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-58506-9 |
| _version_ | 1797199480229986304 |
|---|---|
| author | Thi Thuy Uyen Nguyen Anh-Tien Nguyen Hyeongwan Kim Yu Jin Jung Woong Park Kyoung Min Kim Ilwoo Park Won Kim |
| author_facet | Thi Thuy Uyen Nguyen Anh-Tien Nguyen Hyeongwan Kim Yu Jin Jung Woong Park Kyoung Min Kim Ilwoo Park Won Kim |
| author_sort | Thi Thuy Uyen Nguyen |
| collection | DOAJ |
| description | Abstract Tubular injury is the most common cause of acute kidney injury. Histopathological diagnosis may help distinguish between the different types of acute kidney injury and aid in treatment. To date, a limited number of study has used deep-learning models to assist in the histopathological diagnosis of acute kidney injury. This study aimed to perform histopathological segmentation to identify the four structures of acute renal tubular injury using deep-learning models. A segmentation model was used to classify tubule-specific injuries following cisplatin treatment. A total of 45 whole-slide images with 400 generated patches were used in the segmentation model, and 27,478 annotations were created for four classes: glomerulus, healthy tubules, necrotic tubules, and tubules with casts. A segmentation model was developed using the DeepLabV3 architecture with a MobileNetv3-Large backbone to accurately identify the four histopathological structures associated with acute renal tubular injury in PAS-stained mouse samples. In the segmentation model for four structures, the highest Intersection over Union and the Dice coefficient were obtained for the segmentation of the “glomerulus” class, followed by “necrotic tubules,” “healthy tubules,” and “tubules with cast” classes. The overall performance of the segmentation algorithm for all classes in the test set included an Intersection over Union of 0.7968 and a Dice coefficient of 0.8772. The Dice scores for the glomerulus, healthy tubules, necrotic tubules, and tubules with cast are 91.78 ± 11.09, 87.37 ± 4.02, 88.08 ± 6.83, and 83.64 ± 20.39%, respectively. The utilization of deep learning in a predictive model has demonstrated promising performance in accurately identifying the degree of injured renal tubules. These results may provide new opportunities for the application of the proposed methods to evaluate renal pathology more effectively. |
| first_indexed | 2024-04-24T07:16:25Z |
| format | Article |
| id | doaj.art-879a4f525f8541d1bb80274eab1cdac0 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-24T07:16:25Z |
| publishDate | 2024-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-879a4f525f8541d1bb80274eab1cdac02024-04-21T11:18:29ZengNature PortfolioScientific Reports2045-23222024-04-0114111110.1038/s41598-024-58506-9Deep-learning model for evaluating histopathology of acute renal tubular injuryThi Thuy Uyen Nguyen0Anh-Tien Nguyen1Hyeongwan Kim2Yu Jin Jung3Woong Park4Kyoung Min Kim5Ilwoo Park6Won Kim7Department of Histology, Embryology, Pathology and Forensic Medicine, Hue University of Medicine and Pharmacy, Hue UniversityDepartment of Radiology, Chonnam National University and HospitalDepartment of Internal Medicine, Jeonbuk National University Medical SchoolDepartment of Internal Medicine, Jeonbuk National University Medical SchoolDepartment of Internal Medicine, Jeonbuk National University Medical SchoolDepartment of Pathology, Jeonbuk National University Medical SchoolDepartment of Radiology, Chonnam National University and HospitalDepartment of Internal Medicine, Jeonbuk National University Medical SchoolAbstract Tubular injury is the most common cause of acute kidney injury. Histopathological diagnosis may help distinguish between the different types of acute kidney injury and aid in treatment. To date, a limited number of study has used deep-learning models to assist in the histopathological diagnosis of acute kidney injury. This study aimed to perform histopathological segmentation to identify the four structures of acute renal tubular injury using deep-learning models. A segmentation model was used to classify tubule-specific injuries following cisplatin treatment. A total of 45 whole-slide images with 400 generated patches were used in the segmentation model, and 27,478 annotations were created for four classes: glomerulus, healthy tubules, necrotic tubules, and tubules with casts. A segmentation model was developed using the DeepLabV3 architecture with a MobileNetv3-Large backbone to accurately identify the four histopathological structures associated with acute renal tubular injury in PAS-stained mouse samples. In the segmentation model for four structures, the highest Intersection over Union and the Dice coefficient were obtained for the segmentation of the “glomerulus” class, followed by “necrotic tubules,” “healthy tubules,” and “tubules with cast” classes. The overall performance of the segmentation algorithm for all classes in the test set included an Intersection over Union of 0.7968 and a Dice coefficient of 0.8772. The Dice scores for the glomerulus, healthy tubules, necrotic tubules, and tubules with cast are 91.78 ± 11.09, 87.37 ± 4.02, 88.08 ± 6.83, and 83.64 ± 20.39%, respectively. The utilization of deep learning in a predictive model has demonstrated promising performance in accurately identifying the degree of injured renal tubules. These results may provide new opportunities for the application of the proposed methods to evaluate renal pathology more effectively.https://doi.org/10.1038/s41598-024-58506-9Deep learningAcute renal tubular injurySegmentation |
| spellingShingle | Thi Thuy Uyen Nguyen Anh-Tien Nguyen Hyeongwan Kim Yu Jin Jung Woong Park Kyoung Min Kim Ilwoo Park Won Kim Deep-learning model for evaluating histopathology of acute renal tubular injury Scientific Reports Deep learning Acute renal tubular injury Segmentation |
| title | Deep-learning model for evaluating histopathology of acute renal tubular injury |
| title_full | Deep-learning model for evaluating histopathology of acute renal tubular injury |
| title_fullStr | Deep-learning model for evaluating histopathology of acute renal tubular injury |
| title_full_unstemmed | Deep-learning model for evaluating histopathology of acute renal tubular injury |
| title_short | Deep-learning model for evaluating histopathology of acute renal tubular injury |
| title_sort | deep learning model for evaluating histopathology of acute renal tubular injury |
| topic | Deep learning Acute renal tubular injury Segmentation |
| url | https://doi.org/10.1038/s41598-024-58506-9 |
| work_keys_str_mv | AT thithuyuyennguyen deeplearningmodelforevaluatinghistopathologyofacuterenaltubularinjury AT anhtiennguyen deeplearningmodelforevaluatinghistopathologyofacuterenaltubularinjury AT hyeongwankim deeplearningmodelforevaluatinghistopathologyofacuterenaltubularinjury AT yujinjung deeplearningmodelforevaluatinghistopathologyofacuterenaltubularinjury AT woongpark deeplearningmodelforevaluatinghistopathologyofacuterenaltubularinjury AT kyoungminkim deeplearningmodelforevaluatinghistopathologyofacuterenaltubularinjury AT ilwoopark deeplearningmodelforevaluatinghistopathologyofacuterenaltubularinjury AT wonkim deeplearningmodelforevaluatinghistopathologyofacuterenaltubularinjury |