Guwahati, July 12, 2023: A team of researchers from the Indian Institute of Technology, Guwahati (IIT-G) has developed an Artificial Intelligence (AI)-based model that can predict the severity of knee osteoarthritis from X-ray images. The model, called OsteoHRNet, uses Deep Learning (DL) techniques to automatically assess the condition of the knee joints and help doctors diagnose the disease more accurately.

Knee osteoarthritis is a common and painful disorder that affects the cartilage and bones in the knee. It is estimated that 28% of the Indian population suffers from this disease. The conventional methods of diagnosis, such as MRI and CT scans, are expensive and not widely available. X-ray imaging, on the other hand, is cheaper and more accessible, but requires expert interpretation.

The IIT-G researchers have overcome this challenge by creating OsteoHRNet, which can analyze X-ray images and classify them into different severity levels of knee osteoarthritis. The model can also provide a score that indicates the degree of joint damage. The researchers said that their model can assist medical practitioners remotely and improve the quality of diagnosis and treatment.

The research team was led by Prof. Sur at IIT Guwahati, and included former PhD students Prasen Kumar Sharma and Sibaji Gaj (now a research fellow at Cleveland Clinic, Ohio, USA).