Research Project
Evaluating whether multimodal AI systems can convincingly add disease findings to normal radiological images - and releasing a curated, radiologist-scored synthetic dataset for benchmarking.
Leads: Dr. Shreyas Reddy & Sarath Sasi (CRASH Lab, Ashoka). CRASH Lab lead: Dr. Suvrankar Datta.
Status: Active - Supported by Google.org through Gemini Academic Program.
An ongoing study comparing how the newest reasoning AI models interpret radiology scans against practicing radiologists - and how confident the AI is in its own readings.
Active - Simons Foundation
View research →Creating the first standardised classification of the visual reasoning mistakes AI models make on radiology images - so they can be spotted, measured, and fixed.
Active - Simons Foundation
View research →Creating scalable autonomous pipelines to learn how individual radiologists write their reports, so that AI-generated radiology reports preserve each doctor's reporting voice while staying medically accurate.
Active - Self-Funded
View research →