WGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformer
Abstract The vertebral foramen, lamina, and vertebral body are three critical components of the spine structure, essential for maintaining spinal connectivity and stability. Accurately segmenting lumbar structures such as the vertebral body, vertebral foramen, and lamina in MRI cross-sections helps...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-11-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-79244-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846165357480378368 |
|---|---|
| author | Jing Liu Guodong Suo Fengqing Jin Yuee Zhou Jianlan Yang |
| author_facet | Jing Liu Guodong Suo Fengqing Jin Yuee Zhou Jianlan Yang |
| author_sort | Jing Liu |
| collection | DOAJ |
| description | Abstract The vertebral foramen, lamina, and vertebral body are three critical components of the spine structure, essential for maintaining spinal connectivity and stability. Accurately segmenting lumbar structures such as the vertebral body, vertebral foramen, and lamina in MRI cross-sections helps doctors better understand and diagnose the pathological causes of spine-related diseases. This study presents a multi-structure semantic segmentation method for vertebral transverse section MRI slices using WGAN with a residual U-Net and clustered Transformer. The generator network was replaced with a combination of a residual U-Net and a clustered Transformer-based segmentation network. The enhanced U-Net encoder, utilizing dilated convolutions and residual structures, improved multi-scale feature extraction capabilities. Meanwhile, the clustered Transformer structure, with reduced progressive linear complexity, ensured the extraction of global positional information. The results of multiple experiments show that the Dice coefficient for vertebral body segmentation increased by 3.1%, the Hausdorff distance decreased by 0.6 mm, mIOU improved by 4.1–96.2%, and PPV increased by 2.0–98.8% compared to mainstream segmentation models. These improvements are statistically significant (p < 0.05).Ablation experiments further validated the effectiveness of the proposed enhanced modules in improving segmentation accuracy for the three target structures. |
| format | Article |
| id | doaj-art-cfa39c653c494af7a4bd49fa8b3f8f1c |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-cfa39c653c494af7a4bd49fa8b3f8f1c2024-11-17T12:24:09ZengNature PortfolioScientific Reports2045-23222024-11-0114111810.1038/s41598-024-79244-yWGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformerJing Liu0Guodong Suo1Fengqing Jin2Yuee Zhou3Jianlan Yang4School of Medical Information Engineering, Gansu University of Chinese MedicineSchool of Medical Information Engineering, Gansu University of Chinese MedicineSchool of Medical Information Engineering, Gansu University of Chinese MedicineSchool of Medical Information Engineering, Gansu University of Chinese MedicineOrthopedic Traumatology HospitalAbstract The vertebral foramen, lamina, and vertebral body are three critical components of the spine structure, essential for maintaining spinal connectivity and stability. Accurately segmenting lumbar structures such as the vertebral body, vertebral foramen, and lamina in MRI cross-sections helps doctors better understand and diagnose the pathological causes of spine-related diseases. This study presents a multi-structure semantic segmentation method for vertebral transverse section MRI slices using WGAN with a residual U-Net and clustered Transformer. The generator network was replaced with a combination of a residual U-Net and a clustered Transformer-based segmentation network. The enhanced U-Net encoder, utilizing dilated convolutions and residual structures, improved multi-scale feature extraction capabilities. Meanwhile, the clustered Transformer structure, with reduced progressive linear complexity, ensured the extraction of global positional information. The results of multiple experiments show that the Dice coefficient for vertebral body segmentation increased by 3.1%, the Hausdorff distance decreased by 0.6 mm, mIOU improved by 4.1–96.2%, and PPV increased by 2.0–98.8% compared to mainstream segmentation models. These improvements are statistically significant (p < 0.05).Ablation experiments further validated the effectiveness of the proposed enhanced modules in improving segmentation accuracy for the three target structures.https://doi.org/10.1038/s41598-024-79244-yResidual U-NetClustered transformerLumbar spine MRIWGANMulti-structure |
| spellingShingle | Jing Liu Guodong Suo Fengqing Jin Yuee Zhou Jianlan Yang WGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformer Scientific Reports Residual U-Net Clustered transformer Lumbar spine MRI WGAN Multi-structure |
| title | WGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformer |
| title_full | WGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformer |
| title_fullStr | WGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformer |
| title_full_unstemmed | WGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformer |
| title_short | WGAN-based multi-structure segmentation of vertebral cross-section MRI using ResU-Net and clustered transformer |
| title_sort | wgan based multi structure segmentation of vertebral cross section mri using resu net and clustered transformer |
| topic | Residual U-Net Clustered transformer Lumbar spine MRI WGAN Multi-structure |
| url | https://doi.org/10.1038/s41598-024-79244-y |
| work_keys_str_mv | AT jingliu wganbasedmultistructuresegmentationofvertebralcrosssectionmriusingresunetandclusteredtransformer AT guodongsuo wganbasedmultistructuresegmentationofvertebralcrosssectionmriusingresunetandclusteredtransformer AT fengqingjin wganbasedmultistructuresegmentationofvertebralcrosssectionmriusingresunetandclusteredtransformer AT yueezhou wganbasedmultistructuresegmentationofvertebralcrosssectionmriusingresunetandclusteredtransformer AT jianlanyang wganbasedmultistructuresegmentationofvertebralcrosssectionmriusingresunetandclusteredtransformer |