Deep learning-based automatic detection and grading of disk herniation in lumbar magnetic resonance images
Abstract Magnetic resonance imaging of the lumbar spine is a key technique for clarifying the cause of disease. The greatest challenges today are the repetitive and time-consuming process of interpreting these complex MR images and the problem of unequal diagnostic results from physicians with diffe...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-10401-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849344702956961792 |
|---|---|
| author | Yan Guo Xiaoxiang Huang Wei Chen Ichiro Nakamoto Weiqing Zhuang Hua Chen Jie Feng Jianfeng Wu |
| author_facet | Yan Guo Xiaoxiang Huang Wei Chen Ichiro Nakamoto Weiqing Zhuang Hua Chen Jie Feng Jianfeng Wu |
| author_sort | Yan Guo |
| collection | DOAJ |
| description | Abstract Magnetic resonance imaging of the lumbar spine is a key technique for clarifying the cause of disease. The greatest challenges today are the repetitive and time-consuming process of interpreting these complex MR images and the problem of unequal diagnostic results from physicians with different levels of experience. To address these issues, in this study, an improved YOLOv8 model (GE-YOLOv8) that combines a gradient search module and efficient channel attention was developed. To address the difficulty of intervertebral disc feature extraction, the GS module was introduced into the backbone network, which enhances the feature learning ability for the key structures through the gradient splitting strategy, and the number of parameters was reduced by 2.1%. The ECA module optimizes the weights of the feature channels and enhances the sensitivity of detection for small-target lesions, and the mAP50 was improved by 4.4% compared with that of YOLOv8. GE-YOLOv8 demonstrated the significance of this innovation on the basis of a P value <.001, with YOLOv8 as the baseline. The experimental results on a dataset from the Pingtan Branch of Union Hospital of Fujian Medical University and an external test dataset show that the model has excellent accuracy. |
| format | Article |
| id | doaj-art-050c8012c3314793ba14293cd80f2d5b |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-050c8012c3314793ba14293cd80f2d5b2025-08-20T03:42:36ZengNature PortfolioScientific Reports2045-23222025-07-0115111210.1038/s41598-025-10401-7Deep learning-based automatic detection and grading of disk herniation in lumbar magnetic resonance imagesYan Guo0Xiaoxiang Huang1Wei Chen2Ichiro Nakamoto3Weiqing Zhuang4Hua Chen5Jie Feng6Jianfeng Wu7School of Internet Economics and Business, Fujian University of TechnologySchool of Internet Economics and Business, Fujian University of TechnologySchool of Internet Economics and Business, Fujian University of TechnologySchool of Internet Economics and Business, Fujian University of TechnologySchool of Internet Economics and Business, Fujian University of TechnologyDepartment of Radiology, Pingtan Comprehensive Experimentation Area HospitalDepartment of Radiology, Pingtan Comprehensive Experimentation Area HospitalDepartment of Neurosurgery, Fujian Medical University Union HospitalAbstract Magnetic resonance imaging of the lumbar spine is a key technique for clarifying the cause of disease. The greatest challenges today are the repetitive and time-consuming process of interpreting these complex MR images and the problem of unequal diagnostic results from physicians with different levels of experience. To address these issues, in this study, an improved YOLOv8 model (GE-YOLOv8) that combines a gradient search module and efficient channel attention was developed. To address the difficulty of intervertebral disc feature extraction, the GS module was introduced into the backbone network, which enhances the feature learning ability for the key structures through the gradient splitting strategy, and the number of parameters was reduced by 2.1%. The ECA module optimizes the weights of the feature channels and enhances the sensitivity of detection for small-target lesions, and the mAP50 was improved by 4.4% compared with that of YOLOv8. GE-YOLOv8 demonstrated the significance of this innovation on the basis of a P value <.001, with YOLOv8 as the baseline. The experimental results on a dataset from the Pingtan Branch of Union Hospital of Fujian Medical University and an external test dataset show that the model has excellent accuracy.https://doi.org/10.1038/s41598-025-10401-7Lumbar spine symptomsYOLOv8Magnetic resonance imagingMSUGradient shunt |
| spellingShingle | Yan Guo Xiaoxiang Huang Wei Chen Ichiro Nakamoto Weiqing Zhuang Hua Chen Jie Feng Jianfeng Wu Deep learning-based automatic detection and grading of disk herniation in lumbar magnetic resonance images Scientific Reports Lumbar spine symptoms YOLOv8 Magnetic resonance imaging MSU Gradient shunt |
| title | Deep learning-based automatic detection and grading of disk herniation in lumbar magnetic resonance images |
| title_full | Deep learning-based automatic detection and grading of disk herniation in lumbar magnetic resonance images |
| title_fullStr | Deep learning-based automatic detection and grading of disk herniation in lumbar magnetic resonance images |
| title_full_unstemmed | Deep learning-based automatic detection and grading of disk herniation in lumbar magnetic resonance images |
| title_short | Deep learning-based automatic detection and grading of disk herniation in lumbar magnetic resonance images |
| title_sort | deep learning based automatic detection and grading of disk herniation in lumbar magnetic resonance images |
| topic | Lumbar spine symptoms YOLOv8 Magnetic resonance imaging MSU Gradient shunt |
| url | https://doi.org/10.1038/s41598-025-10401-7 |
| work_keys_str_mv | AT yanguo deeplearningbasedautomaticdetectionandgradingofdiskherniationinlumbarmagneticresonanceimages AT xiaoxianghuang deeplearningbasedautomaticdetectionandgradingofdiskherniationinlumbarmagneticresonanceimages AT weichen deeplearningbasedautomaticdetectionandgradingofdiskherniationinlumbarmagneticresonanceimages AT ichironakamoto deeplearningbasedautomaticdetectionandgradingofdiskherniationinlumbarmagneticresonanceimages AT weiqingzhuang deeplearningbasedautomaticdetectionandgradingofdiskherniationinlumbarmagneticresonanceimages AT huachen deeplearningbasedautomaticdetectionandgradingofdiskherniationinlumbarmagneticresonanceimages AT jiefeng deeplearningbasedautomaticdetectionandgradingofdiskherniationinlumbarmagneticresonanceimages AT jianfengwu deeplearningbasedautomaticdetectionandgradingofdiskherniationinlumbarmagneticresonanceimages |