Decomposition of musculoskeletal structures from radiographs using an improved CycleGAN framework
Abstract This paper presents methods of decomposition of musculoskeletal structures from radiographs into multiple individual muscle and bone structures. While existing solutions require dual-energy scan for the training dataset and are mainly applied to structures with high-intensity contrast, such...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-05-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-023-35075-x |
| _version_ | 1797818073466535936 |
|---|---|
| author | Naoki Nakanishi Yoshito Otake Yuta Hiasa Yi Gu Keisuke Uemura Masaki Takao Nobuhiko Sugano Yoshinobu Sato |
| author_facet | Naoki Nakanishi Yoshito Otake Yuta Hiasa Yi Gu Keisuke Uemura Masaki Takao Nobuhiko Sugano Yoshinobu Sato |
| author_sort | Naoki Nakanishi |
| collection | DOAJ |
| description | Abstract This paper presents methods of decomposition of musculoskeletal structures from radiographs into multiple individual muscle and bone structures. While existing solutions require dual-energy scan for the training dataset and are mainly applied to structures with high-intensity contrast, such as bones, we focused on multiple superimposed muscles with subtle contrast in addition to bones. The decomposition problem is formulated as an image translation problem between (1) a real X-ray image and (2) multiple digitally reconstructed radiographs, each of which contains a single muscle or bone structure, and solved using unpaired training based on the CycleGAN framework. The training dataset was created via automatic computed tomography (CT) segmentation of muscle/bone regions and virtually projecting them with geometric parameters similar to the real X-ray images. Two additional features were incorporated into the CycleGAN framework to achieve a high-resolution and accurate decomposition: hierarchical learning and reconstruction loss with the gradient correlation similarity metric. Furthermore, we introduced a new diagnostic metric for muscle asymmetry directly measured from a plain X-ray image to validate the proposed method. Our simulation and real-image experiments using real X-ray and CT images of 475 patients with hip diseases suggested that each additional feature significantly enhanced the decomposition accuracy. The experiments also evaluated the accuracy of muscle volume ratio measurement, which suggested a potential application to muscle asymmetry assessment from an X-ray image for diagnostic and therapeutic assistance. The improved CycleGAN framework can be applied for investigating the decomposition of musculoskeletal structures from single radiographs. |
| first_indexed | 2024-03-13T09:02:46Z |
| format | Article |
| id | doaj.art-82842923f5cc44cbb29b04847ab494f4 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-13T09:02:46Z |
| publishDate | 2023-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-82842923f5cc44cbb29b04847ab494f42023-05-28T11:15:23ZengNature PortfolioScientific Reports2045-23222023-05-0113111510.1038/s41598-023-35075-xDecomposition of musculoskeletal structures from radiographs using an improved CycleGAN frameworkNaoki Nakanishi0Yoshito Otake1Yuta Hiasa2Yi Gu3Keisuke Uemura4Masaki Takao5Nobuhiko Sugano6Yoshinobu Sato7Division 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 TechnologyDivision 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 Orthopaedic Medical Engineering, Osaka University Graduate School of MedicineDepartment of Bone and Joint Surgery, Ehime University Graduate School of MedicineDepartment of Orthopaedic Medical Engineering, Osaka University Graduate School of MedicineDivision of Information Science, Graduate School of Science and Technology, Nara Institute of Science and TechnologyAbstract This paper presents methods of decomposition of musculoskeletal structures from radiographs into multiple individual muscle and bone structures. While existing solutions require dual-energy scan for the training dataset and are mainly applied to structures with high-intensity contrast, such as bones, we focused on multiple superimposed muscles with subtle contrast in addition to bones. The decomposition problem is formulated as an image translation problem between (1) a real X-ray image and (2) multiple digitally reconstructed radiographs, each of which contains a single muscle or bone structure, and solved using unpaired training based on the CycleGAN framework. The training dataset was created via automatic computed tomography (CT) segmentation of muscle/bone regions and virtually projecting them with geometric parameters similar to the real X-ray images. Two additional features were incorporated into the CycleGAN framework to achieve a high-resolution and accurate decomposition: hierarchical learning and reconstruction loss with the gradient correlation similarity metric. Furthermore, we introduced a new diagnostic metric for muscle asymmetry directly measured from a plain X-ray image to validate the proposed method. Our simulation and real-image experiments using real X-ray and CT images of 475 patients with hip diseases suggested that each additional feature significantly enhanced the decomposition accuracy. The experiments also evaluated the accuracy of muscle volume ratio measurement, which suggested a potential application to muscle asymmetry assessment from an X-ray image for diagnostic and therapeutic assistance. The improved CycleGAN framework can be applied for investigating the decomposition of musculoskeletal structures from single radiographs.https://doi.org/10.1038/s41598-023-35075-x |
| spellingShingle | Naoki Nakanishi Yoshito Otake Yuta Hiasa Yi Gu Keisuke Uemura Masaki Takao Nobuhiko Sugano Yoshinobu Sato Decomposition of musculoskeletal structures from radiographs using an improved CycleGAN framework Scientific Reports |
| title | Decomposition of musculoskeletal structures from radiographs using an improved CycleGAN framework |
| title_full | Decomposition of musculoskeletal structures from radiographs using an improved CycleGAN framework |
| title_fullStr | Decomposition of musculoskeletal structures from radiographs using an improved CycleGAN framework |
| title_full_unstemmed | Decomposition of musculoskeletal structures from radiographs using an improved CycleGAN framework |
| title_short | Decomposition of musculoskeletal structures from radiographs using an improved CycleGAN framework |
| title_sort | decomposition of musculoskeletal structures from radiographs using an improved cyclegan framework |
| url | https://doi.org/10.1038/s41598-023-35075-x |
| work_keys_str_mv | AT naokinakanishi decompositionofmusculoskeletalstructuresfromradiographsusinganimprovedcycleganframework AT yoshitootake decompositionofmusculoskeletalstructuresfromradiographsusinganimprovedcycleganframework AT yutahiasa decompositionofmusculoskeletalstructuresfromradiographsusinganimprovedcycleganframework AT yigu decompositionofmusculoskeletalstructuresfromradiographsusinganimprovedcycleganframework AT keisukeuemura decompositionofmusculoskeletalstructuresfromradiographsusinganimprovedcycleganframework AT masakitakao decompositionofmusculoskeletalstructuresfromradiographsusinganimprovedcycleganframework AT nobuhikosugano decompositionofmusculoskeletalstructuresfromradiographsusinganimprovedcycleganframework AT yoshinobusato decompositionofmusculoskeletalstructuresfromradiographsusinganimprovedcycleganframework |