Tassoker Meleka, Hakli Huseyinb, Yaman Metin Furkanb, Ekmekcı Sema Nurb, Incekara Senanurb, Kamaci Serhatb, Ozturk Busraa
aNecmettin Erbakan University, Faculty of Dentistry, Department of Oral and Maxillofacial Radiology, Konya, Türkiye
bNecmettin Erbakan University, Department of Computer Engineering, Konya, Türkiye
Abstract
Background/Aim: The temporomandibular joint (TMJ) is a complex anatomical region composed of the mandibular condyle located in the glenoid fossa of the temporal bone and covered with fibrous connective tissue. Excessive and continuous forces lead to progressive degeneration of the bony surfaces of the TMJ. The aim of this study is to determine the success of automatic detection of degenerative changes detected on panoramic radiographs in the TMJ region with deep learning method. Material and Methods: Panoramic images of 1068 patients (1000 with normal TMJ appearance and 68 with TMJ degeneration) over 18 years of age were included in the study. CVAT, open-source annotation tool (https://www.cvat.ai/) was used for labeling image data. All images were resized using the bilinear interpolation method. With the using data augmentation techniques, the number of images data reached 1480. BSRGAN model was applied to the data to increase the resolution of the data. YOLOv5, YOLOv7 and YOLOv8 algorithms were used for TMJ degeneration detection. TP, FP, TN, FN, accuracy, precision, recall, F1-score and AUC (Area Under the Curve) metrics were used for statistical analysis. Results: YOLOv5s training resulted in 94.40% accuracy, 81.63% precision, 86.96% sensitivity, 84.21% F1 score and 91.45% AUC. YOLOv7 training resulted in 99.63% accuracy, 97.87% precision, 100% sensitivity, 98.92% F1 Score and 99.77% AUC. YOLOv8 training resulted 96.64% accuracy, 91.11% precision, 89.13% sensitivity, 90.11% F1 Score and 93.66% AUC. Conclusions: All three algorithms have high success rates, with the best results obtained in YOLOv7.
Keywords: deep learning; degeneration; panoramic radiography; temporomandibular joint
Reference
Bäck, K., Ahlqwist, M., Hakeberg, M., Dahlström, L. (2017) Occurrence of signs of osteoarthritis/arthrosis in the temporomandibular joint on panoramic radiographs in Swedish women. Community Dent Oral Epidemiol, 45(5): 478-484
Bochkovskiy, A., Wang, C.Y., Liao, H. (2020) Yolov4: Optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934
Buckwalter, J.A., Mankin, H.J., Grodzinsky, A.J. (2005) Articular cartilage and osteoarthritis. Instr Course Lect, 54: 465-480
Çelik, Ş. (2022) Application of Artificial Neural Networks in Different Activation Functions: Prediction of Duck Population in Turkey. OKU J Inst Sci Technol, 5 (2): 800-811
Choi, E., Kim, D., Lee, J.Y., Park, H.K. (2021) Artificial intelligence in detecting temporomandibular joint osteoarthritis on orthopantomogram. Sci Rep, 11 (1): 10246
Dsouza, V.D., Rao, P.K., Kini, R. (2020) Assessment of degenerative changes in the temporomandibular joint: A retrospective cone-beam computed tomography study. SRM J Res Dent Sci, 11 (4): 195-198
Erzurumlu, Z.Ü., Çelenk, P. (2022) Effects of aging on temporomandibular joint. Turk Dent Res J, 1(2): 75-78
Eşer, G., Duman, Ş.B., Bayrakdar, İ.Ş., Çelik, Ö. (2023) Classification of temporomandibular joint osteoarthritis on cone beam computed tomography images using artificial intelligence system. J Oral Rehabil, 50(9): 758-766
Göceri, E. (2023) Medical image data augmentation: Techniques, comparisons and interpretations. Artif Intell Rev, 56 (11): 12561-12605
Güney, E. (2021) Mobile gpu based real-time situation analysis and detection applications for driver assistance systems. Sakarya University
Jocher, G., Nishimura, K., Mineeva, T., Vilariño, R., Yolov (2024) Yolov5 Code repository. https://github.com/ultralytics/ yolov5.2020
Jung, W., Lee, K.E., Suh, B.J., Seok, H., Lee, D.W. (2023) Deep learning for osteoarthritis classification in temporomandibular joint. Oral Dis, 29 (3): 1050-1059
Kaya, E., Gunec, H.G., Gokyay, S.S., Kutal, S., Gulum, S., Ates, H.F. (2022) Proposing a CNN Method for Primary and Permanent Tooth Detection and Enumeration on Pediatric Dental Radiographs. J Clin Pediatr Dent, 46(4): 293-293
Kim, D., Choi, E., Jeong, H.G., Chang, J., Youm, S. (2020) Expert System for Mandibular Condyle Detection and Osteoarthritis Classification in Panoramic Imaging Using R-CNN and CNN. Appl Sci (Basel), 10: 7464
Kivrak, O., Gürbüz, M.Z. (2022) Performance comparison of YOLOv3, YOLOv4 and YOLOv5 algorithms: A case study for poultry recognition. EJOSAT, 38: 392-397
Koç, N. (2020) Evaluation of osteoarthritic changes in the temporomandibular joint and their correlations with age: A retrospective CBCT study. Dent Med Probl, 57 (1): 67-72
Krishnamoorthy, V.K., Baskaran, S. (2022) Optimized adversarial network with faster residual deep learning for osteoarthritis classification in panoramic radiography. International Journal of Intelligent Engineering and Systems, 15(6): 191-200
Kulkarni, S., Seneviratne, N., Baig, M.S., Khan, A.H.A. (2020) Artificial intelligence in medicine: Where are we now?. Acad Radiol, 27(1): 62-70
Li, K., Wang, Y., Hu, Z. (2023) Improved YOLOv7 for small object detection algorithm based on attention and dynamic convolution. Appl Sci (Basel), 13 (16): 9316
Li, M., Punithakumar, K., Major, P.W., Nguyen, K.C.T., Pereira, C.P. (2022) Temporomandibular joint segmentation in MRI images using deep learning. J Dent, 127: 104345
Okeson, J. (2019) Management of temporomandibular disorders and occlusion. USA: Mosbey, 8th ed
Orhan, K., Driesen, L., Shujaat, S., Jacobs, R., Chai, X. (2021) Development and Validation of a Magnetic Resonance Imaging-Based Machine Learning Model for TMJ Pathologies. Biomed Res Int, 2021: 6656773
Raghuram, S., Bharadwaj, A.S., Deepika, S.K., Khadabadi, M.S., Jayaprakash, A.A. (2022) Digital Implementation of the Softmax Activation Function and the Inverse Softmax Function. u: 4th International Conference on Circuits, Control, Communication and Computing, 64-67
1
Redmon, J., Farhadi, A. (2018) Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767
Redmon, J., Farhadi, A. (2017) YOLO9000: Better, Faster, Stronger. u: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 6517-6525
1
Redmon, J., Divvala, S., Girshick, R., Farhadi, A. (2016) You only look once: Unified, real-time object detection. u: Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 779-788
Roboflow (n.d) What is YOLOv7? A Complete Guide. https://blog.roboflow.com/yolov7breakdown
Serindere, G., Belgin, C.A. (2019) Examination of condylar changes with panoramic radiography. Selcuk Dent J, 6(4): 173-177
Shetty, U.S., Burde, K.N., Naikmasur, V.G., Sattur, A.P. (2014) Assessment of condylar changes in patients with temporomandibular joint pain using digital volumetric tomography. Radiol Res Pract, 2014: 106059
Terven, J., Córdova-Esparza, D.M., Romero-González, J.A.A. (2023) Comprehensive review of YOLO architectures in computer vision: From YOLOv1 to YOLOv8 and YOLONAS. Mach Learn Knowl Extr, 5 (4): 1680-1716
Titus, J., Geroge, S. (2013) A comparison study on different interpolation methods based on satellite images. Int J Eng Res Technol, 2(6): 82-85
Wang, C., Bochkovskiy, Y.A., Liao, H.Y.M. (2022) YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for realtime object detectors. u: Proc IEEE Conf Comput Vis Pattern Recognit
Wang, H., Zhang, S., Zhao, S., Wang, Q., Li, D., Zhao, R. (2022) Real-time detection and tracking of fish abnormal behavior based on improved YOLOV5 and SiamRPN. Comput Electron Agric, 192: 106512
Wang, X., Zhang, J.N., Gan, Y.H., Zhou, Y.H. (2015) Current understanding of pathogenesis and treatment of TMJ osteoarthritis. J Dent Res, 94 (5): 666-673
Yan, Y., Chen, M., Shyu, M., Chen, S. (2015) Deep learning for imbalanced multimedia data classification. u: IEEE International Symposium on Multimedia (ISM), Miami, Fl, USA: IEEE, 483-488
Yıldırım, D., Alkış, Ü. (2016) Imaging methods used in the evaluation of temporomandibular joint disorders. Med J SDU, 7(2)
Yıldız, G., Yıldız, D. (2018) Brain tumour location and tumour area calculation by means of morphological processes and edge detection methods. IJMSIT, 2 (2): 39-42
Zhang, K., Liang, J., van Gool, L., Timofte, R. (2021) Designing a practical degradation model for deep blind image super-resolution. u: IEEE/CVF International Conference on Computer Vision (ICCV), IEEE, 4771-4780
Balkan Journal of Dental Medicine, 2024, vol. 28, br. 2, str. 99-116