Wednesday, Nov 18, 2026
English News
  • Hyderabad
  • Telangana
  • AP News
  • India
  • World
  • Entertainment
  • Sport
  • Science and Tech
  • Business
  • Rewind
  • ...
    • NRI
    • View Point
    • cartoon
    • My Space
    • Education Today
    • Reviews
    • Property
    • Lifestyle
E-Paper
  • NRI
  • View Point
  • cartoon
  • My Space
  • Reviews
  • Education Today
  • Property
  • Lifestyle
Home | Hyderabad | Iiit Hyderabad Study Cautions Against Blindly Trusting Medical Ai Models

IIIT-Hyderabad study cautions against blindly trusting medical AI models

Researchers at IIIT-Hyderabad have cautioned against blindly trusting medical AI models after auditing four vision-language models on chest X-rays. The study found that diagnostic information could influence image localisation, while radiologists preferred broader areas around abnormalities in some cases.

By Telangana Today
Updated On - 8 October 2026, 07:01 PM
IIIT-Hyderabad study cautions against blindly trusting medical AI models
whatsapp facebook twitter telegram

Hyderabad: Medical AI tools are increasingly being used in medical imaging, where radiologists rely on them as a second pair of eyes to flag possible abnormalities in scans. However, a study by International Institute of Information Technology (IIIT) – Hyderabad researchers cautioned against blindly trusting output from such medical AI models.

The research team led by Prof. Parameswari Krishnamurthy, with Dr. Syed Faizan as principal investigator, at the Language Technologies Research Centre (LTRC), audited four vision-language models — MAIRA-2, MedGemma-4B, LLaVA-Med-1.5 and LLaVA-1.5 — on thousands of publicly available chest X-rays.


The team also conducted a reader study in which two radiologists assessed anonymised overlays, providing a human reference for the models’ localisation.

According to Dr. Faizan, an AI model may appear to highlight the correct part of an image, but that does not necessarily mean it has identified the disease in the same way a radiologist would. “The model may first arrive at a diagnosis and then use that diagnosis to determine where to place its heat map. In other words, it may be working backwards,” he said.

To test this possibility, the team removed the diagnostic information and examined what happened to the models’ localisation. The models’ performance dropped, suggesting that some of what looked like image-based reasoning could actually be influenced by an anatomical expectation learned from the diagnosis.

The researchers also had a 2-radiologist study, like a human-in-the-loop, to calibrate their findings and discovered that they rated MedGemma higher even though the audit had ranked Myra to be superior in overlapping heatmaps.

“This suggests that a model paying attention to a particular area – which may be the exact spot where the disorder lies – is not exactly helpful to a radiologist. A radiologist might want to look at not only areas with the disease, but also a broader surrounding area to know the extent of disease spread. MedGemma is doing that; it provides a broader area,” he explained.

The study was accepted at MICCAI 2026 (International Conference on Medical Image Computing and Computer Assisted Intervention) in Strasbourg at the iMIMIC Satellite Event.

The LTRC is also studying how reliably medical vision-language models respond when doctors phrase the same question differently. In collaboration with a radiologist from CMC Vellore, researchers examined the models’ responses to different paraphrases of medical questions, with the study accepted at EMNLP.

Join Our Whatsapp Channel
  • Follow Us :
As preferred source
  • Tags
  • IIIT Hyderabad
  • MAIRA-2
  • medical AI models
  • medical imaging AI

Related News

  • Royal Enfield, Wikimedia Foundation and IIIT-H join hands to preserve Himalayan heritage

    Royal Enfield, Wikimedia Foundation and IIIT-H join hands to preserve Himalayan heritage

  • IIIT Hyderabad launches advanced BrainLab for human brain research

    IIIT Hyderabad launches advanced BrainLab for human brain research

  • IIIT Hyderabad, Hexagon R&D India set up Digital Twin Centre of Excellence

    IIIT Hyderabad, Hexagon R&D India set up Digital Twin Centre of Excellence

  • IIIT-H launches Centre for Educational Technology and Learning Sciences

    IIIT-H launches Centre for Educational Technology and Learning Sciences

Latest News

  • Sourav Ganguly backs Rishabh Pant’s return to Delhi Capitals

    2 minutes ago
  • Mahabubabad MP Balram Naik’s controversial remarks draws criticism on social media

    16 minutes ago
  • Hyderabad’s SFR Snehit wins men’s singles title at National Ranking Table Tennis Championships

    38 minutes ago
  • Editorial: Countering China’s dominance

    42 minutes ago
  • Opinion: India deserves a seat at the APEC table

    43 minutes ago
  • VIT ranks 293rd globally, fifth in India in QS Sustainability Rankings 2027

    52 minutes ago
  • Sreenidi Deccan FC face FC Bengaluru United in crucial IFL clash

    1 hour ago
  • Amee Kamani begins IBSF World Snooker Championship campaign with twin wins

    1 hour ago

company

  • Home
  • About Us
  • Contact Us
  • Privacy Policy

business

  • Subscribe

telangana today

  • Telangana
  • Hyderabad
  • Latest News
  • Entertainment
  • World
  • Andhra Pradesh
  • Science & Tech
  • Sport

follow us

  • Telangana Today Telangana Today
Telangana Today Telangana Today

© Copyrights 2024 TELANGANA PUBLICATIONS PVT. LTD. All rights reserved. Powered by Veegam

Telangana Today App