IISc and Aster CMI Hospital develop AI tool to help in nerve disorder screening

The Indian Institute of Science (IISc) and Aster-CMI Hospital jointly created an AI tool for detecting carpal tunnel syndrome (CTS) by identifying the median nerve in ultrasound videos.

CTS occurs when the median nerve, compressed at the wrist’s carpal tunnel, leads to numbness and pain, often affecting those with repetitive hand movements. While ultrasound is used for assessment, interpreting the images can be challenging, explained Karan R. Gujarati, first author and former MTech student at the Department of Computational and Data Sciences (CDS), IISc.

The team employed a transformer-based machine learning model, akin to ChatGPT’s architecture, modifying it to focus on tracking the median nerve.

• Also Read: Samsung, IISc sign MoU to set up quantum technology lab

The researchers at IISc collaborated with Lokesh Bathala, Lead Consultant Neurologist at Aster-CMI Hospital, to collect and annotate ultrasound videos from both healthy participants and people with CTS to train the model.

The model, functioning like tracking an object in a video, could segment the median nerve and measure its cross-sectional area in real time, aiding CTS diagnosis with over 95 per cent accuracy at the wrist. Unlike models for CT and MRI scans, there’s a scarcity of ultrasound video-focused models, especially for nerves.

The tool, initially trained on one nerve, aims to expand to all upper and lower limb nerves. A successful pilot test at the hospital involved real-time monitoring on an additional display connected to an ultrasound machine.

• Also Read: IISc’s Materials Research Centre engineers enzyme mimetic to break down waste under sunlight

Going forward, the team envisions integrating the tool into ultrasound machines, potentially reducing diagnosis time for doctors. However, they emphasize that a physician’s final diagnosis remains crucial.

The lead neurologist anticipates collaboration with ultrasound machine manufacturers for wider implementation, highlighting the tool’s ability to provide real-time assistance during examinations. The overall goal is to enhance diagnostic efficiency, acknowledging the indispensable role of medical professionals in the final decision-making process.

COMMENT NOW