Deep learning-assisted screening and diagnosis of scoliosis: segmentation of bare-back images via an attention-enhanced convolutional neural network
Abstract Background Traditional diagnostic tools for scoliosis screening necessitate a substantial number of specialized personnel and equipment, leading to inconvenience that can result in missed opportunities for early diagnosis and optimal treatment. We have developed a deep learning-based image...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-02-01
|
| Series: | Journal of Orthopaedic Surgery and Research |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13018-025-05564-y |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850197189651857408 |
|---|---|
| author | Xingyu Duan Xiaojun Ma Mengqi Zhu Linan Wang Dingqi You Lili Deng Ningkui Niu |
| author_facet | Xingyu Duan Xiaojun Ma Mengqi Zhu Linan Wang Dingqi You Lili Deng Ningkui Niu |
| author_sort | Xingyu Duan |
| collection | DOAJ |
| description | Abstract Background Traditional diagnostic tools for scoliosis screening necessitate a substantial number of specialized personnel and equipment, leading to inconvenience that can result in missed opportunities for early diagnosis and optimal treatment. We have developed a deep learning-based image segmentation model to enhance the efficiency of scoliosis screening. Methods A total of 350 patients with scoliosis and 108 healthy subjects were included in this study. The dataset was created using their bare back images and standing full-length anteroposterior spinal X-rays. An attention mechanism was incorporated into the original U-Net architecture to build a Dual AttentionUNet model for image segmentation. The entire dataset was divided into the training (321 cases), validation (46 cases), and test (91 cases) sets in a 7:1:2 ratio. The training set was used to train the Dual AttentionUNet model, and the validation set was used to fine-tune hyperparameters and prevent overfitting during training. The performance of the model was evaluated in the test set. After automatic segmentation of the back contour, a back asymmetry index was calculated via computer vision algorithms to classify scoliosis into different severities. The accuracy of classifications was statistically compared to those of three clinical experts. Results Following the segmentation of bare back images and the application of computer vision algorithms, the Dual AttentionUNet model achieved an accuracy, precision, and recall rate of over 90% in predicting severe scoliosis. Notably, the model achieved an AUC value of 0.93 in identifying whether the subjects had scoliosis, which was higher than the 0.92 achieved by the deputy chief physician. In identifying severe scoliosis, their AUC values were 0.95 and 0.96, respectively. Conclusion The Dual AttentionUNet model, based on only bare back images, achieved accuracy and precision comparable to clinical physicians in determining scoliosis severity. Radiation-free, cost-saving, easy-to-operate and noninvasive, this model provides a novel option for large-scale scoliosis screening. |
| format | Article |
| id | doaj-art-301ee67afbfd4e4282ee7af616c9b319 |
| institution | OA Journals |
| issn | 1749-799X |
| language | English |
| publishDate | 2025-02-01 |
| publisher | BMC |
| record_format | Article |
| series | Journal of Orthopaedic Surgery and Research |
| spelling | doaj-art-301ee67afbfd4e4282ee7af616c9b3192025-08-20T02:13:14ZengBMCJournal of Orthopaedic Surgery and Research1749-799X2025-02-0120111610.1186/s13018-025-05564-yDeep learning-assisted screening and diagnosis of scoliosis: segmentation of bare-back images via an attention-enhanced convolutional neural networkXingyu Duan0Xiaojun Ma1Mengqi Zhu2Linan Wang3Dingqi You4Lili Deng5Ningkui Niu6Department of Orthopedics, General Hospital of Ningxia Medical UniversityDepartment of Orthopedics, General Hospital of Ningxia Medical UniversityDepartment of Orthopedics, General Hospital of Ningxia Medical UniversityDepartment of Orthopedics, General Hospital of Ningxia Medical UniversityDepartment of Spinal Cord Surgery, Henan Provincial People’s HospitalDepartment of General Practice, Zhengzhou First People’s HospitalDepartment of Orthopedics, General Hospital of Ningxia Medical UniversityAbstract Background Traditional diagnostic tools for scoliosis screening necessitate a substantial number of specialized personnel and equipment, leading to inconvenience that can result in missed opportunities for early diagnosis and optimal treatment. We have developed a deep learning-based image segmentation model to enhance the efficiency of scoliosis screening. Methods A total of 350 patients with scoliosis and 108 healthy subjects were included in this study. The dataset was created using their bare back images and standing full-length anteroposterior spinal X-rays. An attention mechanism was incorporated into the original U-Net architecture to build a Dual AttentionUNet model for image segmentation. The entire dataset was divided into the training (321 cases), validation (46 cases), and test (91 cases) sets in a 7:1:2 ratio. The training set was used to train the Dual AttentionUNet model, and the validation set was used to fine-tune hyperparameters and prevent overfitting during training. The performance of the model was evaluated in the test set. After automatic segmentation of the back contour, a back asymmetry index was calculated via computer vision algorithms to classify scoliosis into different severities. The accuracy of classifications was statistically compared to those of three clinical experts. Results Following the segmentation of bare back images and the application of computer vision algorithms, the Dual AttentionUNet model achieved an accuracy, precision, and recall rate of over 90% in predicting severe scoliosis. Notably, the model achieved an AUC value of 0.93 in identifying whether the subjects had scoliosis, which was higher than the 0.92 achieved by the deputy chief physician. In identifying severe scoliosis, their AUC values were 0.95 and 0.96, respectively. Conclusion The Dual AttentionUNet model, based on only bare back images, achieved accuracy and precision comparable to clinical physicians in determining scoliosis severity. Radiation-free, cost-saving, easy-to-operate and noninvasive, this model provides a novel option for large-scale scoliosis screening.https://doi.org/10.1186/s13018-025-05564-yScoliosisArtificial intelligenceDeep learningImage segmentationScreening |
| spellingShingle | Xingyu Duan Xiaojun Ma Mengqi Zhu Linan Wang Dingqi You Lili Deng Ningkui Niu Deep learning-assisted screening and diagnosis of scoliosis: segmentation of bare-back images via an attention-enhanced convolutional neural network Journal of Orthopaedic Surgery and Research Scoliosis Artificial intelligence Deep learning Image segmentation Screening |
| title | Deep learning-assisted screening and diagnosis of scoliosis: segmentation of bare-back images via an attention-enhanced convolutional neural network |
| title_full | Deep learning-assisted screening and diagnosis of scoliosis: segmentation of bare-back images via an attention-enhanced convolutional neural network |
| title_fullStr | Deep learning-assisted screening and diagnosis of scoliosis: segmentation of bare-back images via an attention-enhanced convolutional neural network |
| title_full_unstemmed | Deep learning-assisted screening and diagnosis of scoliosis: segmentation of bare-back images via an attention-enhanced convolutional neural network |
| title_short | Deep learning-assisted screening and diagnosis of scoliosis: segmentation of bare-back images via an attention-enhanced convolutional neural network |
| title_sort | deep learning assisted screening and diagnosis of scoliosis segmentation of bare back images via an attention enhanced convolutional neural network |
| topic | Scoliosis Artificial intelligence Deep learning Image segmentation Screening |
| url | https://doi.org/10.1186/s13018-025-05564-y |
| work_keys_str_mv | AT xingyuduan deeplearningassistedscreeninganddiagnosisofscoliosissegmentationofbarebackimagesviaanattentionenhancedconvolutionalneuralnetwork AT xiaojunma deeplearningassistedscreeninganddiagnosisofscoliosissegmentationofbarebackimagesviaanattentionenhancedconvolutionalneuralnetwork AT mengqizhu deeplearningassistedscreeninganddiagnosisofscoliosissegmentationofbarebackimagesviaanattentionenhancedconvolutionalneuralnetwork AT linanwang deeplearningassistedscreeninganddiagnosisofscoliosissegmentationofbarebackimagesviaanattentionenhancedconvolutionalneuralnetwork AT dingqiyou deeplearningassistedscreeninganddiagnosisofscoliosissegmentationofbarebackimagesviaanattentionenhancedconvolutionalneuralnetwork AT lilideng deeplearningassistedscreeninganddiagnosisofscoliosissegmentationofbarebackimagesviaanattentionenhancedconvolutionalneuralnetwork AT ningkuiniu deeplearningassistedscreeninganddiagnosisofscoliosissegmentationofbarebackimagesviaanattentionenhancedconvolutionalneuralnetwork |