IIT-G develops deep learning-based framework to assess knee OA severity from X-ray images

The AI-based model, named called OsteoHRNet, can be used to detect the severity level of OA and assist medical practitioners remotely for a more accurate diagnosis, a release by IIT, Guwahati, said.Knee osteoarthritis is the most common musculoskeletal disorder in the world.


PTI | Guwahati | Updated: 10-07-2023 15:35 IST | Created: 10-07-2023 15:35 IST
IIT-G develops deep learning-based framework to assess knee OA severity from X-ray images
  • Country:
  • India

Indian Institute of Technology (IIT), Guwahati, has developed a deep learning (DL)-based framework that automatically assesses the severity of knee osteoarthritis (OA) from X-ray images. The AI-based model, named called OsteoHRNet, can be used to detect the severity level of OA and assist medical practitioners remotely for a more accurate diagnosis, a release by IIT, Guwahati, said.

Knee osteoarthritis is the most common musculoskeletal disorder in the world. Its prevalence is a high 28 per cent in India. As there is no cure for it but total joint replacement at an advanced stage, an early diagnosis is essential for pain management and behavioral corrections, it said.

Researchers of the institute have been working to enhance automatic knee osteoarthritis detection from X-Ray images or radiographs to assist clinical evaluation. Their team developed the AI-based model which automatically assesses the severity of knee OA, the release said.

Arijit Sur, a professor in the computer science and engineering department of IIT-Guwahati, said, "The proposed model may be a good starting point for analyzing inexpensive radiographic modalities such as X-rays. Our group is currently focusing on how efficient deep learning based models can be designed so that we can work on inexpensive and easily available modalities such as very low-resolution radiographic images or even photos taken from radiographic plates by a smartphone." MRI and CT scans provide 3-D image of knee joints for effective diagnosis of the disorder. Their availability is, however, limited and expensive. X-ray imaging is very affective for routine diagnosis and more economically feasible. Palash Ghosh, assistant professor in the department of mathematics in the institute, said, "Compared to the others that exist, the model developed by us can pinpoint the area which is medically most important to decide the severity level of knee osteoarthritis. It can help medical practitioners detect the disorder accurately at an early stage".

The research by Rohit Kumar Jain under the joint supervision of Sur and Ghosh, has been accepted for publication in the journal 'Multimedia Tools and Applications'. The research team also included Prasen Kumar Sharma and Sibaji Gaj. The model predicts knee OA severity according to the World Health Organization approved Kellgren and Lawrence (KL) grading scale that ranges from grade 0 (low severity) to 4 (high severity), the release said.

It is built on one of the most recent deep models called the high-resolution network (HRNet) to capture the multi-scale features of knee X-rays. It uses an efficient deep convolutional neural network, an algorithm from image recognition. It is not a direct plug-and-play of the popular dep models, the release said.

"The team is working to reconfigure these models in such a way that they can be deployed in resource-constrained devices so that medical professionals can easily get an initial but accurate guess for the diagnosis. ''This work has the potential to mitigate the severe shortage of skilled personnel in this field, especially in rural India," it added.

(This story has not been edited by Devdiscourse staff and is auto-generated from a syndicated feed.)

Give Feedback