Article Details

Abstract

Accurate imaging evaluation of Anterior Cruciate Ligament (ACL) injuries of the knee joint is crucial for clinical diagnosis and treatment. Due to the slender anatomical structure of the ACL, its small volume, and complex surrounding tissues, traditional Deep Learning models face difficulties in balancing global dependencies with local details. This study proposes an automatic segmentation model based on SwinUNet-FPN. The model uses Swin-UNet with a pure Transformer architecture as the backbone to extract global features through a shifted window self-attention mechanism and introduces a Feature Pyramid Network (FPN) module to fuse deep semantic information with shallow details through lateral connections. Experimental results on the KneeMRI dataset show that the model achieves Dice Similarity Coefficient (DSC) and 95% Hausdorff Distance (HD95) of 0.812 7 and 18.234 5, respectively, significantly outperforming U-Net, ResUNet, and the original Swin-UNet models. The results confirm that this method can effectively overcome segmentation challenges in complex backgrounds, improve edge recognition accuracy, and provide important clinical auxiliary value.

Keywords

anterior cruciate ligament; MRI image segmentation; Swin-UNet; Feature Pyramid Network (FPN); Deep Learning

References

1. [1] Chia L，De O S D，Whalan M，et al. Non-Contact Anterior Cruciate Ligament Injury Epidemiology in Team-Ball Sports: A Systematic Review With Meta-Analysis by Sex, Age, Sport, Participation Level, and Exposure Type [J].Sports Medicine (Auckland，N.Z.)，2022，52（10）：2447-2467.
2. [2] Wang Yong, Yu Zeping, Li Jian, etc Analysis of risk factors for contralateral anterior cruciate ligament injury after initial anterior cruciate ligament reconstruction surgery [J]. Chinese Journal of Reconstructive Surgery, 2023, 37 (5): 578-583
3. [3] Templeton K. Musculoskeletal Disorders: Sex and Gender Evidence in Anterior Cruciate Ligament Injuries, Osteoarthritis, and Osteoporosis [M]//Jenkins M, Newman C B. How Sex and Gender Impact Clinical Practice: An Evidence-Based Guide to Patient Care，2021：207-227.
4. [4] Zhao M G，Zhou Y，Chang J R，et al. The Accuracy of MRI in the Diagnosis of Anterior Cruciate Ligament Injury [J/OL].Annals of Translational Medicine，2020，8（24）：1657.https://atm.amegroups.org/article/view/58843/html.
5. [5] Almajalid R，Zhang M，Shan J. Fully Automatic Knee Bone Detection and Segmentation on Three-Dimensional MRI [J].Diagnostics，2022，12（1）：123.https://doi.org/10.3390/diagnostics12010123.
6. [6] Yusuke M，Ryuichiro A，Satoshi M，et al. Automated Detection of Anterior Cruciate Ligament Tears Using a Deep Convolutional Neural Network [J/OL].BMC Musculoskeletal Disorders，2022，23（1）：577.https://link.springer.com/article/10.1186/s12891-022-05524-1.
7. [7] Long J，Shelhamer E，Darrell T. Fully Convolutional Networks for Semantic Segmentation [C]//IEEE Conference on Computer Vision and Pattern Recognition，2015：3431-3440.
8. [8] Milletari F，Navab N，Ahmadi S A. V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation [C]//2016 Fourth International Conference on 3D Vision (3DV)，2016：565-571.
9. [9] Ronneberger O，Fischer P，Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation [C]//Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015，2015：234-241.
10. [10] Flannery S W， Kiapour A M， Edgar D J，et al. Automated Magnetic Resonance Image Segmentation of the Anterior Cruciate Ligament [J].Journal of Orthopaedic Research：Official Publication of the Orthopaedic Research Society，2020，39（4）：1-10.
11. [11] Awan M J，Rahim M S M，Salim N，et al. Automated Knee MR Images Segmentation of Anterior Cruciate Ligament Tears [J/OL].Sensors，2022，22（4）：1552.https://doi.org/10.3390/s22041552.
12. [12] Dung N T，Thuan N H，Dung T V，et al. End-to-End Deep Learning Model for Segmentation and Severity Staging of Anterior Cruciate Ligament Injuries From MRI [J].Diagnostic and Interventional Imaging，2023，104（3）：133-141.
13. [13] Zhang Z，Liu Q，Wang Y. Road Extraction by Deep Residual U-Net [J].IEEE Geoscience and Remote Sensing Letters，2018，15：749-753.
14. [14] Chen J N，Mei J R，Li X H，et al. TransUNet: Rethinking the U-Net Architecture Design for Medical Image Segmentation Through the Lens of Transformers [J/OL].Medical Image Analysis，2024，97：103280.https://doi.org/10.1016/j.media.2024.103280.
15. [15] Cao H U，Wang Y Y，Chen J，et al. Swin-UNet: UNET-Like Pure Transformer for Medical Image Segmentation [C]//European Conference on Computer Vision，2022：205-218.

Funding

Guangxi Higher Education Undergraduate Teaching Reform Project (2023JGZ115); Key R&D Program Project in Guangxi (Guike AB25069086)

About the authors

Lu Zhenzhan (2001-), male, Zhuang ethnicity, from Hechi, Guangxi. He is a master's student with research interests in medical image segmentation and medical artificial intelligence; Ren Wenning (1999-), male, Han ethnicity, from Nanning, Guangxi, is a master's student with research interests in deep learning algorithms and medical informatics; Corresponding author: He Huimin (1973-), female, Han ethnicity, from Miluo, Hunan, professor, doctoral student, research direction: medical information management, medical artificial intelligence.