An efficient classification of knee Osteoarthritis severity grading using SqueezeNet model

View at publisher Summarise with VIRAT
Authors: Nagarani Nagarajan, Anandan Malaiarasan, P. Rajeswari, Sivasankari Jothiraj, B. Muneeswari, Janani Govindaraj
Cited by: 0
References: 10

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Knee osteoarthritis (KOA) is a time honored musculoskeletal condition characterized via deterioration of the cartilage in the knee joint. It regularly leads to signs and symptoms such as ache, stiffness, swelling, and reduced mobility. KOA can profoundly affect each day activities such as Sports, taking walks, hiking stairs, and bending, thereby diminishing ordinary exceptional of life for the once affected. Determining the course of a disease and guiding treatment recommendations require timely and precise grading decisions. Our proposed method leverages the efficiency and effectiveness of SqueezeNet to extract discriminative qualities from knee radiographs. Based on rigorous testing on a large dataset with different degrees of OA severity, our method performs well in terms of accurately estimating knee Osteoarthritisseverity. Our approach demonstrates high performance in accurately estimating the severity of knee osteoarthritis. The utilization of knee images, the model undergoes training and evaluation using the SqueezeNet algorithm for osteoarthritis grading, achieving a remarkable accuracy of 91%. The results demonstrate how deep learning, and specifically SqueezeNet, can enhance the customized treatment plans and diagnostic processes for knee OA patients.

Fields of Study & Subjects

Medical Imaging and Analysis Artificial Intelligence in Healthcare and Education Osteoarthritis Treatment and Mechanisms

Publication details

Journal / source
AIP conference proceedings
Volume
3341
Pages
020058–020058
Publication date
1 January 2026
Language
EN