Osteoarthritis Classification Using MobileNetV3 Model by Data Pre-processing and Depiction on Fine - Tuned Optimized Parameters

• Published in: 2023 Global Conference on Information Technologies and Communications (GCITC) pp. 1 - 4
• Main Authors: Gill, Kanwarpartap Singh, Anand, Vatsala, Gupta, Rupesh
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2023
• Subjects: CNN; Confusion matrix; Convolutional neural networks; Data models; Deep Learning; Joints; Mobile handsets; MobileNetV2 Model; Model Training; Osteoarthritis; Predictive models; Public healthcare
• ISBN: 9798350308143
• DOI: 10.1109/GCITC60406.2023.10426151

A frequent degenerative joint illness that damages the cartilage, the protective tissue that covers the ends of bones inside a joint, is osteoarthritis, sometimes known as OA. It can result in discomfort, stiffness, swelling, and a reduction in the range of motion in the joints that are afflicted. Osteoarthritis may affect any joint, including the hands, feet, and shoulders, although it most frequently affects weight-bearing joints like the knees, hips, and spine. Age, obesity, joint damage, repeated joint usage, genetic susceptibility, and other underlying medical issues such metabolic disorders are all risk factors for developing osteoarthritis. The categorization of osteoarthritis in this research is done using MobileNetV3 neural network architecture. In applications where computing resources are constrained, such as on mobile phones or other edge devices, MobileNetV3 is a lightweight convolutional neural network (CNN) architecture that is created to be effective and suited for deployment on mobile devices. To effectively fight osteoarthritis in its beginning stages, it is important to first determine its type or category. Our MobileNetV3 model is very good at classifying osteoarthritis, achieving an accuracy rate of over 80%.