This paper is published in Volume-11, Issue-1, 2025
Area
Engineering
Author
Usha Kumari V, Abhishek S, Ajay R, Karnan K, Asim Ulla Khan
Org/Univ
Acharya Institute of Technology, Karnataka, India
Pub. Date
04 February, 2025
Paper ID
V11I1-1318
Publisher
Keywords
Knee Osteoarthritis, CenterNet, Object Detection, Pixel-Wise Voting, Medical Imaging, Deep Learning, Convolutional Neural Networks (CNN)

Citationsacebook

IEEE
Usha Kumari V, Abhishek S, Ajay R, Karnan K, Asim Ulla Khan. Refined Detection of Knee Osteoarthritis Using Center Net with a Pixel-Wise Voting Approach, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
Usha Kumari V, Abhishek S, Ajay R, Karnan K, Asim Ulla Khan (2025). Refined Detection of Knee Osteoarthritis Using Center Net with a Pixel-Wise Voting Approach. International Journal of Advance Research, Ideas and Innovations in Technology, 11(1) www.IJARIIT.com.

MLA
Usha Kumari V, Abhishek S, Ajay R, Karnan K, Asim Ulla Khan. "Refined Detection of Knee Osteoarthritis Using Center Net with a Pixel-Wise Voting Approach." International Journal of Advance Research, Ideas and Innovations in Technology 11.1 (2025). www.IJARIIT.com.

Abstract

This paper introduces an advanced approach for detecting Knee Osteoarthritis (OA) using an optimized CenterNet framework integrated with a pixel-wise voting strategy. Early and accurate detection of knee OA is vital for timely intervention and efficient disease management. The proposed method enhances the CenterNet architecture—a leading object detection framework—by incorporating a pixel-based voting mechanism, which leverages local image data to improve detection accuracy. Each pixel contributes to determining whether it belongs to an object or the background, and this aggregated information enables precise identification of objects and their locations. Experiments conducted on a publicly available knee OA dataset demonstrate that the proposed method outperforms existing techniques, achieving state-of-the-art results. The integration of CenterNet with the pixel-wise voting strategy holds significant promise in aiding clinicians with early diagnosis and treatment planning for knee OA patients.