(Universitas Internasional Batam;
Universiti Teknikal Malaysia Melaka, Malaysia)(2) * Yogan Jaya Kumar
(Universiti Teknikal Malaysia Melaka, Malaysia)(3) Sek Yong Wee
(Universiti Teknikal Malaysia Melaka, Malaysia)(4) Vinod Kumar Perhakaran
(Universiti Tunku Abdul Rahman, Malaysia)*corresponding author
AbstractThe accuracy of diagnosing an Anterior Cruciate Ligament (ACL) tear depends on the radiologist’s or surgeon’s expertise, experience, and skills. In this study, we contribute to the development of an automated diagnostic model for anterior cruciate ligament (ACL) tears using a lightweight deep learning model, specifically ResNet-14, combined with a Spatial Attention mechanism to enhance diagnostic performance while conserving computational resources. The model processes knee MRI scans using a ResNet architecture, comprising a series of residual blocks and a spatial attention mechanism, to focus on the essential features in the imaging data. The methodology, which includes the training and evaluation process, was conducted using the Stanford dataset, comprising 1,370 knee MRI scans. Data augmentation techniques were also implemented to mitigate biases. The model’s assessment uses performance metrics, ROC-AUC, sensitivity, and specificity. The results show that the proposed model achieved an ROC-AUC score of 0.8696, a sensitivity of 79.81%, and a specificity of 79.82%. At 6.67 MB in size, with 1,684,517 parameters, the model is significantly more compact than existing models, such as MRNet. The findings demonstrate that embedding spatial attention into a lightweight deep learning framework augments the diagnostic accuracy for ACL tears while maintaining computational efficiency. Therefore, lightweight models have the potential to enhance diagnostic capability in medical imaging, allowing them to be deployed in resource-constrained clinical settings.
KeywordsAnterior cruciate ligament; ResNet-14; Spatial attention mechanism; MRNet; Lightweight deep learning
|
DOIhttps://doi.org/10.26555/ijain.v11i3.2055 |
Article metricsAbstract views : 2099 | PDF views : 176 |
Cite |
Full Text Download
|
References
[1] L. Andriollo et al., “The Role of Artificial Intelligence in Anterior Cruciate Ligament Injuries: Current Concepts and Future Perspectives,” Healthc., vol. 12, no. 3, pp. 1–16, 2024, doi: 10.3390/healthcare12030300.
[2] R. Crawford, G. Walley, S. Bridgman, and N. Maffulli, “Magnetic resonance imaging versus arthroscopy in the diagnosis of knee pathology, concentrating on meniscal lesions and ACL tears: A systematic review,” Br. Med. Bull., vol. 84, no. 1, pp. 5–23, 2007, doi: 10.1093/bmb/ldm022.
[3] C. Murmu, P. Tiwari, S. Sircar, and V. Agrawal, “Accuracy of magnetic resonance imaging in diagnosis of knee injuries,” Int. J. Orthop. Sci., vol. 3, no. 1b, pp. 85–88, 2017, doi: 10.22271/ortho.2017.v3.i1b.15.
[4] Z. Li et al., “Deep Learning-Based Magnetic Resonance Imaging Image Features for Diagnosis of Anterior Cruciate Ligament Injury,” J. Healthc. Eng., vol. 2021, 2021, doi: 10.1155/2021/4076175.
[5] S. S. Mazlan, M. Z. Ayob, and Z. A. Kadir Bakti, “Anterior cruciate ligament (ACL) injury classification system using support vector machine (SVM),” in 2017 International Conference on Engineering Technology and Technopreneurship (ICE2T), Sep. 2017, no. September, pp. 1–5, doi: 10.1109/ICE2T.2017.8215960.
[6] C. I. Girard et al., “Decision Tree Learning Algorithm for Classifying Knee Injury Status Using Return-to-Activity Criteria,” Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBS, vol. 2020-July, pp. 5494–5497, 2020, doi: 10.1109/EMBC44109.2020.9176010.
[7] M. Martyna, W. Tomasz, G. J. Krzysztof, and G. Monika, “Artificial neural networks in knee injury risk evaluation among professional football players,” in AIP Conference Proceedings, 2018, vol. 1922, no. February 2006, p. 070002, doi: 10.1063/1.5019069.
[8] I. Štajduhar, M. Mamula, D. Miletić, and G. Ünal, “Semi-automated detection of anterior cruciate ligament injury from MRI,” Comput. Methods Programs Biomed., vol. 140, pp. 151–164, 2017, doi: 10.1016/j.cmpb.2016.12.006.
[9] Q. Zhang, Y. Xiao, J. Yang, F. Deng, Z. Zhang, and J. Cai, “The value of 2D and 3D MRI texture models in Grade II and III anterior cruciate ligament injuries,” Knee, vol. 54, no. 87, pp. 254–262, 2025, doi: 10.1016/j.knee.2025.01.007.
[10] Y. S. Jeon et al., “Interpretable and Lightweight 3-D Deep Learning Model for Automated ACL Diagnosis,” IEEE J. Biomed. Heal. Informatics, vol. 25, no. 7, pp. 2388–2397, 2021, doi: 10.1109/JBHI.2021.3081355.
[11] W. Hu and S. Razmjooy, “Combining an improved political optimizer with convolutional neural networks for accurate anterior cruciate ligament tear detection in sports injuries,” Sci. Rep., vol. 15, no. 1, pp. 1–14, 2025, doi: 10.1038/s41598-025-91242-2.
[12] K. Joshi and K. Suganthi, “Anterior cruciate ligament tear detection based on convolutional neural network and generative adversarial neural network,” Neural Comput. Appl., vol. 36, no. 9, pp. 5021–5030, 2023, doi: 10.1007/s00521-023-09350-x.
[13] J. A. J. Alsayaydeh, M. F. bin Yusof, M. Z. B. A. Halim, M. N. S. Zainudin, and S. G. Herawan, “Patient Health Monitoring System Development using ESP8266 and Arduino with IoT Platform,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 4, pp. 617–624, 2023, doi: 10.14569/IJACSA.2023.0140467.
[14] C. Wang, K. Huang, and J. Chen, “Lightweight Deep Learning : An Overview,” IEEE Consum. Electron. Mag., vol. 13, no. August, pp. 51–64, 2024, doi: 10.1109/MCE.2022.3181759.
[15] D. Neogi, M. A. Chowdhury, M. M. Akter, and M. I. A. Hossain, “Mobile detection of cataracts with an optimised lightweight deep Edge Intelligent technique,” IET Cyber-Physical Syst. Theory Appl., no. July 2023, pp. 269–281, 2024, doi: 10.1049/cps2.12083.
[16] F. Chen, S. Li, J. Han, F. Ren, and Z. Yang, “Review of Lightweight Deep Convolutional Neural Networks,” Arch. Comput. Methods Eng., vol. 31, no. 4, pp. 1915–1937, 2024, doi: 10.1007/s11831-023-10032-z.
[17] Z. Zhu et al., “Lightweight medical image segmentation network with multi-scale feature-guided fusion,” Comput. Biol. Med., vol. 182, no. September, p. 109204, Nov. 2024, doi: 10.1016/j.compbiomed.2024.109204.
[18] Anthony, Herman, and A. Yulianto, “Development of License Plate Recognition System Introduction Literature review,” J. Ilm. KOMPUTASI, vol. 23, pp. 571–578, 2024. [Online]. Available at: https://ejournal.jak-stik.ac.id/index.php/komputasi/article/view/3659/900.
[19] O. Ukwandu, H. Hindy, and E. Ukwandu, “An evaluation of lightweight deep learning techniques in medical imaging for high precision COVID-19 diagnostics,” Healthc. Anal., vol. 2, no. August, p. 100096, 2022, doi: 10.1016/j.health.2022.100096.
[20] Z. Waheed et al., “A novel lightweight deep learning based approaches for the automatic diagnosis of gastrointestinal disease using image processing and knowledge distillation techniques,” Comput. Methods Programs Biomed., vol. 260, no. December 2024, p. 108579, 2025, doi: 10.1016/j.cmpb.2024.108579.
[21] A. Elaraby, A. Saad, H. Elmannai, M. Alabdulhafith, M. Hadjouni, and M. Hamdi, “An approach for classification of breast cancer using lightweight deep convolution neural network,” Heliyon, vol. 10, no. 20, p. e38524, 2024, doi: 10.1016/j.heliyon.2024.e38524.
[22] H. Kalita, S. Dandapat, and P. Kumar Bora, “Contextual Self-Attention Based UNet Architecture for Fluid Segmentation in Retinal OCT B-scans,” in Proceedings of the Fifteenth Indian Conference on Computer Vision Graphics and Image Processing, Dec. 2024, pp. 1–8, doi: 10.1145/3702250.3702291.
[23] Y. Zhou, H. Wang, H. Jin, Y. Liu, X. Liu, and Z. Cao, “Remaining useful life prediction for machinery using multimodal interactive attention spatial–temporal networks with deep ensembles,” Expert Syst. Appl., vol. 263, no. September 2024, p. 125808, 2025, doi: 10.1016/j.eswa.2024.125808.
[24] J. Zhang et al., “Advances in attention mechanisms for medical image segmentation,” Comput. Sci. Rev., vol. 56, no. December 2024, p. 100721, May 2025, doi: 10.1016/j.cosrev.2024.100721.
[25] C. Salmi, A. Lebcir, A. M. Djemmal, A. Lebcir, and N. Boubendir, “A Machine Learning Model for Automation of Ligament Injury Detection Process,” in Model and Data Engineering: 9th International Conference, MEDI 2019, Toulouse, France, October 28–31, 2019, Proceedings, Springer, 2019, pp. 317–332, doi: 10.1007/978-3-030-32065-2_22.
[26] Y. Zuo, J. Shao, and N. Razmjooy, “Anterior cruciate ligament tear detection using gated recurrent unit and flexible fitness dependent optimizer,” Biomed. Signal Process. Control, vol. 96, no. PB, p. 106616, 2024, doi: 10.1016/j.bspc.2024.106616.
[27] M. G. Ragab et al., “A Comprehensive Systematic Review of YOLO for Medical Object Detection (2018 to 2023),” IEEE Access, vol. 12, no. February, pp. 57815–57836, 2024, doi: 10.1109/ACCESS.2024.3386826.
[28] N. Bien et al., “Deep-learning-assisted diagnosis for knee magnetic resonance imaging: Development and retrospective validation of MRNet,” PLoS Med., vol. 15, no. 11, pp. 1–19, 2018, doi: 10.1371/journal.pmed.1002699.
[29] F. Liu et al., “Fully Automated Diagnosis of Anterior Cruciate Ligament Tears on Knee MR Images by Using Deep Learning,” Radiol. Artif. Intell., vol. 1, no. 3, p. 180091, May 2019, doi: 10.1148/ryai.2019180091.
[30] J. Qiao, “The Application of Artificial Intelligence in Football Risk Prediction,” Comput. Intell. Neurosci., vol. 2022, pp. 1–9, Jun. 2022, doi: 10.1155/2022/6996134.
[31] M. N. Hasan, M. A. Hossain, and M. A. Rahman, “An ensemble based lightweight deep learning model for the prediction of cardiovascular diseases from electrocardiogram images,” Eng. Appl. Artif. Intell., vol. 141, no. December 2024, p. 109782, 2025, doi: 10.1016/j.engappai.2024.109782.
[32] H. Samma, A. S. Bin Sama, and Q. S. Hamad, “Optimized deep learning model for medical image diagnosis,” J. Eng. Res., no. November, pp. 1–9, Nov. 2024, doi: 10.1016/j.jer.2024.11.003.
[33] P. Sharma et al., “A lightweight deep learning model for automatic segmentation and analysis of ophthalmic images,” Sci. Rep., vol. 12, no. 1, pp. 1–18, 2022, doi: 10.1038/s41598-022-12486-w.
[34] O. E. Okman, M. G. Ulkar, and G. S. Uyanik, “L^3U-net: Low-Latency Lightweight U-net Based Image Segmentation Model for Parallel CNN Processors,” in Proceedings of tinyML Research Symposium (tinyML Research Symposium’22), 2022, vol. 1, no. 1, pp. 1–6, [Online]. Available at: https://arxiv.org/abs/2203.16528.
[35] Y. Li, E. Chouzenoux, B. Charmettant, B. Benatsou, J. Lamarque, and N. Lassau, “Lightweight U-Net For Lesion Segmentation In Ultrasound Images,” in 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), Apr. 2021, pp. 611–615, doi: 10.1109/ISBI48211.2021.9434086.
[36] C. Lin, X. Hu, Y. Zhan, and X. Hao, “MobileNetV2 with Spatial Attention module for traffic congestion recognition in surveillance images,” Expert Syst. Appl., vol. 255, no. PD, p. 124701, 2024, doi: 10.1016/j.eswa.2024.124701.
[37] L. Zhang et al., “A novel MRI-based deep learning networks combined with attention mechanism for predicting CDKN2A/B homozygous deletion status in IDH-mutant astrocytoma,” Eur. Radiol., vol. 34, no. 1, pp. 391–399, 2024, doi: 10.1007/s00330-023-09944-y.
[38] X. Zhou et al., “Attention mechanism based multi-sequence MRI fusion improves prediction of response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer,” Radiat. Oncol., vol. 18, no. 1, pp. 1–11, 2023, doi: 10.1186/s13014-023-02352-y.
[39] C. Liang et al., “Effective automatic detection of anterior cruciate ligament injury using convolutional neural network with two attention mechanism modules,” BMC Med. Imaging, vol. 23, no. 1, pp. 1–13, 2023, doi: 10.1186/s12880-023-01091-6.
[40] L. Dai, J. Liu, and Z. Ju, “Binocular Feature Fusion and Spatial Attention Mechanism Based Gaze Tracking,” IEEE Trans. Human-Machine Syst., vol. 52, no. 2, pp. 302–311, Apr. 2022, doi: 10.1109/THMS.2022.3145097.
[41] J. Liao et al., “A machine learning-based feature extraction method for image classification using ResNet architecture,” Digit. Signal Process., vol. 160, no. February, p. 105036, May 2025, doi: 10.1016/j.dsp.2025.105036.
[42] M. Yang, X. Huang, L. Huang, and G. Cai, “Diagnosis of Parkinson’s disease based on 3D ResNet: The frontal lobe is crucial,” Biomed. Signal Process. Control, vol. 85, no. August 2022, p. 104904, Aug. 2023, doi: 10.1016/j.bspc.2023.104904.
[43] S. Benbakreti, M. Benouis, A. Roumane, and S. Benbakreti, “Impact of the data augmentation on the detection of brain tumor from MRI images based on CNN and pretrained models,” Multimed. Tools Appl., vol. 83, no. 13, pp. 39459–39478, 2024, doi: 10.1007/s11042-023-17092-0.
[44] M. Javed Awan, M. Mohd Rahim, N. Salim, M. Mohammed, B. Garcia-Zapirain, and K. Abdulkareem, “Efficient Detection of Knee Anterior Cruciate Ligament from Magnetic Resonance Imaging Using Deep Learning Approach,” Diagnostics, vol. 11, no. 1, p. 105, Jan. 2021, doi: 10.3390/diagnostics11010105.
[45] D. Pan, J. Shen, Z. Al-Huda, and M. A. A. Al-qaness, “VcaNet: Vision Transformer with fusion channel and spatial attention module for 3D brain tumor segmentation,” Comput. Biol. Med., vol. 186, no. September 2024, p. 109662, 2025, doi: 10.1016/j.compbiomed.2025.109662.
[46] X. Yu et al., “An improved medical image segmentation framework with Channel-Height-Width-Spatial attention module,” Eng. Appl. Artif. Intell., vol. 136, no. PA, p. 108751, 2024, doi: 10.1016/j.engappai.2024.108751.
[47] B. J. Erickson and F. Kitamura, “Magician’s corner: 9. performance metrics for machine learning models,” Radiol. Artif. Intell., vol. 3, no. 3, pp. 1–7, 2021, doi: 10.1148/ryai.2021200126.
[48] J. H. Sim, N. Ma’muriyah, and A. Yulianto, “Evaluating YOLOv5 and YOLOv8 : Advancements in Human Detection,” J. Inf. Syst. Informatics, vol. 6, no. 4, pp. 2999–3015, 2024, doi: 10.51519/journalisi.v6i4.944.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
___________________________________________________________
International Journal of Advances in Intelligent Informatics
ISSN 2442-6571 (print) | 2548-3161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
W: http://ijain.org
E: info@ijain.org (paper handling issues)
andri.pranolo.id@ieee.org (publication issues)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0