Validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip-to-knee clinical CT images
Abstract Deep learning-based image segmentation has allowed for the fully automated, accurate, and rapid analysis of musculoskeletal (MSK) structures from medical images. However, current approaches were either applied only to 2D cross-sectional images, addressed few structures, or were validated on...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-024-83793-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559672310464512 |
|---|---|
| author | Mazen Soufi Yoshito Otake Makoto Iwasa Keisuke Uemura Tomoki Hakotani Masahiro Hashimoto Yoshitake Yamada Minoru Yamada Yoichi Yokoyama Masahiro Jinzaki Suzushi Kusano Masaki Takao Seiji Okada Nobuhiko Sugano Yoshinobu Sato |
| author_facet | Mazen Soufi Yoshito Otake Makoto Iwasa Keisuke Uemura Tomoki Hakotani Masahiro Hashimoto Yoshitake Yamada Minoru Yamada Yoichi Yokoyama Masahiro Jinzaki Suzushi Kusano Masaki Takao Seiji Okada Nobuhiko Sugano Yoshinobu Sato |
| author_sort | Mazen Soufi |
| collection | DOAJ |
| description | Abstract Deep learning-based image segmentation has allowed for the fully automated, accurate, and rapid analysis of musculoskeletal (MSK) structures from medical images. However, current approaches were either applied only to 2D cross-sectional images, addressed few structures, or were validated on small datasets, which limit the application in large-scale databases. This study aimed to validate an improved deep learning model for volumetric MSK segmentation of the hip and thigh with uncertainty estimation from clinical computed tomography (CT) images. Databases of CT images from multiple manufacturers/scanners, disease status, and patient positioning were used. The segmentation accuracy, and accuracy in estimating the structures volume and density, i.e., mean HU, were evaluated. An approach for segmentation failure detection based on predictive uncertainty was also investigated. The model has improved all segmentation accuracy and structure volume/density evaluation metrics compared to a shallower baseline model with a smaller training database (N = 20). The predictive uncertainty yielded large areas under the receiver operating characteristic (AUROC) curves (AUROCs ≥ .95) in detecting inaccurate and failed segmentations. Furthermore, the study has shown an impact of the disease severity status on the model’s predictive uncertainties when applied to a large-scale database. The high segmentation and muscle volume/density estimation accuracy and the high accuracy in failure detection based on the predictive uncertainty exhibited the model’s reliability for analyzing individual MSK structures in large-scale CT databases. |
| format | Article |
| id | doaj-art-7dc83fc2134c4cd8afeb5d7ed59bb9d4 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-7dc83fc2134c4cd8afeb5d7ed59bb9d42025-01-05T12:16:17ZengNature PortfolioScientific Reports2045-23222025-01-0115111810.1038/s41598-024-83793-7Validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip-to-knee clinical CT imagesMazen Soufi0Yoshito Otake1Makoto Iwasa2Keisuke Uemura3Tomoki Hakotani4Masahiro Hashimoto5Yoshitake Yamada6Minoru Yamada7Yoichi Yokoyama8Masahiro Jinzaki9Suzushi Kusano10Masaki Takao11Seiji Okada12Nobuhiko Sugano13Yoshinobu Sato14Division of Information Science, Graduate School of Science and Technology, Nara Institute of Science and TechnologyDivision of Information Science, Graduate School of Science and Technology, Nara Institute of Science and TechnologyDepartment of Orthopedic Surgery, Graduate School of Medicine, Osaka UniversityDepartment of Orthopedic Surgery, Graduate School of Medicine, Osaka UniversityDivision of Information Science, Graduate School of Science and Technology, Nara Institute of Science and TechnologyDepartment of Radiology, Keio University School of MedicineDepartment of Radiology, Keio University School of MedicineDepartment of Radiology, Keio University School of MedicineDepartment of Radiology, Keio University School of MedicineDepartment of Radiology, Keio University School of MedicineHitachi Health Care Center, Hitachi Ltd.Department of Bone and Joint Surgery, Graduate School of Medicine, Ehime UniversityDepartment of Orthopedic Surgery, Graduate School of Medicine, Osaka UniversityDepartment of Orthopaedic Medical Engineering, Graduate School of Medicine, Osaka UniversityDivision of Information Science, Graduate School of Science and Technology, Nara Institute of Science and TechnologyAbstract Deep learning-based image segmentation has allowed for the fully automated, accurate, and rapid analysis of musculoskeletal (MSK) structures from medical images. However, current approaches were either applied only to 2D cross-sectional images, addressed few structures, or were validated on small datasets, which limit the application in large-scale databases. This study aimed to validate an improved deep learning model for volumetric MSK segmentation of the hip and thigh with uncertainty estimation from clinical computed tomography (CT) images. Databases of CT images from multiple manufacturers/scanners, disease status, and patient positioning were used. The segmentation accuracy, and accuracy in estimating the structures volume and density, i.e., mean HU, were evaluated. An approach for segmentation failure detection based on predictive uncertainty was also investigated. The model has improved all segmentation accuracy and structure volume/density evaluation metrics compared to a shallower baseline model with a smaller training database (N = 20). The predictive uncertainty yielded large areas under the receiver operating characteristic (AUROC) curves (AUROCs ≥ .95) in detecting inaccurate and failed segmentations. Furthermore, the study has shown an impact of the disease severity status on the model’s predictive uncertainties when applied to a large-scale database. The high segmentation and muscle volume/density estimation accuracy and the high accuracy in failure detection based on the predictive uncertainty exhibited the model’s reliability for analyzing individual MSK structures in large-scale CT databases.https://doi.org/10.1038/s41598-024-83793-7 |
| spellingShingle | Mazen Soufi Yoshito Otake Makoto Iwasa Keisuke Uemura Tomoki Hakotani Masahiro Hashimoto Yoshitake Yamada Minoru Yamada Yoichi Yokoyama Masahiro Jinzaki Suzushi Kusano Masaki Takao Seiji Okada Nobuhiko Sugano Yoshinobu Sato Validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip-to-knee clinical CT images Scientific Reports |
| title | Validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip-to-knee clinical CT images |
| title_full | Validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip-to-knee clinical CT images |
| title_fullStr | Validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip-to-knee clinical CT images |
| title_full_unstemmed | Validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip-to-knee clinical CT images |
| title_short | Validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip-to-knee clinical CT images |
| title_sort | validation of musculoskeletal segmentation model with uncertainty estimation for bone and muscle assessment in hip to knee clinical ct images |
| url | https://doi.org/10.1038/s41598-024-83793-7 |
| work_keys_str_mv | AT mazensoufi validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT yoshitootake validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT makotoiwasa validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT keisukeuemura validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT tomokihakotani validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT masahirohashimoto validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT yoshitakeyamada validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT minoruyamada validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT yoichiyokoyama validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT masahirojinzaki validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT suzushikusano validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT masakitakao validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT seijiokada validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT nobuhikosugano validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages AT yoshinobusato validationofmusculoskeletalsegmentationmodelwithuncertaintyestimationforboneandmuscleassessmentinhiptokneeclinicalctimages |