IIT Guwahati Team Develops Model To Detect Knee Osteoarthritis From X-Ray Images

GUWAHATI: A team of researchers from the Indian Institute of Technology (IIT) Guwahati has successfully developed a groundbreaking model named OsteoHRNet to assess the severity of Knee Osteoarthritis (OA) from X-ray images. This latest discovery of IIT-G is a Deep Learning- based framework that makes use of artificial intelligence (AI) to study X-ray images.

The groundbreaking OsteoHRNet can accurately determine the severity level of the disease and help medical practitioners to remotely diagnose patients with such knee problems.

A statement from IIT Guwahati said that Knee osteoarthritis is the most common musculoskeletal disorder in the world and that India has a prevalence of 28% of the disease. There is no possibility for curing knee OA except through total joint replacement when it reaches an advanced stage.

An early diagnosis is, therefore, essential for pain management and behavioural corrections, it added.

The statement also said that MRI and CT scans can give a 3D image of the knee joints, which is necessary for effective diagnosis of knee OA but technology is limited and expensive, adding that X-ray imaging is very effective for routine diagnosis and more economically feasible.

In order to increase the outcome of the automatic detection of knee osteoarthritis based on X-ray images or radiographs, IIT Guwahati researchers concentrated their efforts on developing an AI-based model which can assess the severity of Knee OA.

An efficient deep convolutional neural network (CNN), which is an algorithm from image recognition, is used by the AI-based model. This model is capable of predicting the severity of knee OA according to the World Health Organisation approved Kellgren and Lawrence (KL) grading scale. It is built upon one of the most recent deep models, called the high-resolution network (HRNet), to capture the multiscale features of knee X-rays

Assistant Professor in the Department of Mathematics at IIT Guwahati, Palash Ghosh, informed that when compared to other approaches, their model can identify the medically crucial area more accurately to determine the severity level of knee osteoarthritis, which can enable medical practitioners to detect the disease accurately at an early stage, thus eliminating the need for joint replacement.

The research at IIT-G was conducted by Rohit Kumar Jain, an MTech Data Science graduate, under the joint supervision of Prof. Arijit Sur and Palash Ghosh. The model has been accepted for publication in the journal Multimedia Tools and Applications. The research team is also comprised of former IIT Guwahati PhD students Prasen Kumar Sharma and Sibaji Gaj.

Also Read: Heavy Downpour In North India Leaves 18 Dead, Several Roads Cut Off

Also Watch: