New AI model diagnoses diseases by drawing visual maps

Scientists have developed a new artificial intelligence (AI) model that accurately identifies tumours and diseases in medical images by explaining each diagnosis with a visual map. The unique transparency of the model, described in the journal IEEE Transactions on Medical Imaging, allows doctors to easily follow its line of reasoning, double-check for accuracy, and explain the results to patients, the researchers said.

Also Read: Flipkart Launches UPI Service in Partnership With Axis Bank

Advancing Early Disease Detection through Transparent AI Models

Sourya Sengupta, the lead author of a study conducted at the Beckman Institute for Advanced Science and Technology in the US, explained the aim of their project: detecting cancer and diseases in their early stages, like to marking an X on a map. Sengupta emphasized that the model’s transparency facilitates easy understanding by doctors, aiding in decision-making explanation to patients. The potential impact is particularly significant in regions with a shortage of medical professionals, where AI can serve as a valuable assistive tool, not a replacement, to expedite the medical image screening process amidst high demand for time and talent.

Also Read: Vyjayanthimala had a Meeting with Prime Minister Narendra Modi

Enhancing Medical Image Review with Transparent AI Models

Rather than replacing human evaluation, an AI model can conduct preliminary scans of medical images, highlighting abnormalities such as tumors or early disease indicators (biomarkers) for subsequent review by doctors. This process not only saves time but can also enhance the interpreting performance of the person analyzing the scan. However, existing models lack transparency in explaining their decisions, prompting the development of a new AI model that self-interprets each decision, providing detailed explanations instead of binary classifications like “tumor versus non-tumor.” The researchers trained this model on over 20,000 images across three different disease diagnosis tasks.

Also Read: Authorities Issue Dress Code as Abu Dhabi Temple Opens to the Public