19. APPLICATION OF MACHINE LEARNING IN DIAGNOSING JAW FRACTURES THROUGH CT SCANNERS: PERFORMANCE ANALYSIS WITH DIFERENT PARAMETERS
Main Article Content
Abstract
Objective: Apply Teachable Machine to detect jaw fractures in CT images.
Subjects and methods: Retrospective study on 1341 images extracted from CT.
Results: Among 746 images with jaw fracture injuries, correct identification occurred at a rate of 91.8% with parameter settings of 50-16-0.001, which decreased gradually to 82.4% when parameter levels were increased to 150:64:0.003. In the mixed set of 1341 images (with and without jaw fractures), the correct identification rate for images with jaw fractures was 87.3% at parameter levels of 50:16:0.001, decreasing to 78.7% when parameters were increased to 150:64:0.003. This demonstrates a correlation between the adjustment of parameter groups such as Epochs, Batch size, and Learning rate to achieve optimal performance, significantly improving accuracy and general prediction ability on data, while avoiding overfitting.
Article Details
Keywords
Artificial intelligence, Teachable Machine, jaw fractures
References
[2] Trần Văn Trường, Trương Mạnh Dũng, Tình hình chấn thương hàm mặt tại Viện Răng Hàm Mặt Hà Nội trong 11 năm (1988-1998), Tạp chí Y học Việt Nam, 1999, 10, tr. 71-80.
[3] Kalmet P.H et al, Deep learning in fracture detection: a narrative review. Acta orthopaedica, 2020, 91 (2), p. 215-220.
[4] Chamunyonga C et al, The impact of artificial intelligence and Teachable Machine in radiation therapy: considerations for future curriculum enhancement, Journal of Medical Imaging and Radiation Sciences, 2020, 51 (2): p. 214-220.
[5] Gulshan V et al, Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs, Jama, 2016, 316 (22), p. 2402-2410.
[6] Olczak J et al, Artificial intelligence for analyzing orthopedic trauma radiographs: deep learning algorithms-are they on par with humans for diagnosing fractures? Acta orthopaedica, 2017, 88 (6), p. 581-586.
[7] Szegedy C et al, Rethinking the Inception Architecture for Computer Vision. In Proceedings of the IEEE conference on computer vision and pattern recognition, June 2016, DOI:10.1109/CVPR.2016.308.
[8] Warin K et al, Assessment of deep convolutional neural network models for mandibular fracture detection in panoramic radiographs, International Journal of Oral and Maxillofacial Surgery, 2022, 51 (11), p. 1488-1494.
[9] Li X et al, Tooth-marked tongue recognition using multiple instance learning and CNN features, IEEE transactions on cybernetics, 2018, 49 (2) p. 380-387.