Suvrankar Datta - Ashoka University

Suvrankar Datta

  • user
  • User logout
Ashoka University

Other links:

Other links:

Suvrankar Datta

Simons-Ashoka Early Career Fellow, Koita Centre for Digital Health – Ashoka (KCDH-A)

M.D. (Radiodiagnosis), AIIMS Delhi
suvrankar.datta@ashoka.edu.in

Dr Suvrankar Datta is a physician-scientist and radiologist trained at AIIMS New Delhi, with an MBBS and a Gold Medal from JIPMER Puducherry. He is a Simons-Ashoka Early Career Fellow at the Koita Centre for Digital Health, and his research group at CRASH Lab develops benchmarks, datasets and evaluation frameworks for safe and reliable clinical AI. His research focuses on evaluating frontier multimodal and language models under clinically realistic conditions, with particular attention to uncertainty, reliability, safe human handover, multilingual performance, bias and real-world workflow fit.

His recent work includes leading Radiology’s Last Exam (RadLE) 1.0 and 2.0, which evaluates frontier AI systems against radiologists and measures not only diagnostic accuracy, but also confidence, safety and readiness to defer to human specialists. He also leads and supervises multi-institutional programmes spanning culturally congruent medical foundation models, multilingual ambient clinical AI and clinical decision support for Indian patients. He is a co-author of multiple frontier AI publications. Through CRASH Lab, he collaborates with clinical, academic, public-sector and technical partners to strengthen responsible AI evaluation and deployment in India and other resource-constrained health systems.

Peer-reviewed journal articles

  • Zhou HY, Acosta JN, Adithan S, Datta S, Topol EJ, Rajpurkar P. MedVersa: A Generalist Foundation Model for Diverse Medical Imaging Tasks. NEJM AI. 2026;3(4).
  • Datta S, Sarangi PK. From Chatbots to Agentic Workflows: Ensuring Responsible Deployment of Large Language Models in Radiology. Indian Journal of Radiology and Imaging. 2026;36(2):286-288. Published online 21 August 2025.
  • Sarangi PK, Mondal H, Datta S. Reply to Comments on “Evaluating ChatGPT-4's Performance in Identifying Radiological Anatomy in FRCR Part 1 Examination Questions.” Indian Journal of Radiology and Imaging. Published online 16 February 2026.
  • Sarangi PK, Datta S, Panda BB, Panda S, Mondal H. Evaluating ChatGPT-4's Performance in Identifying Radiological Anatomy in FRCR Part 1 Examination Questions. Indian Journal of Radiology and Imaging. 2025;35(2):287-294. Published online 4 November 2024.
  • Sarangi PK, Datta S, Swarup MS, Panda S, Nayak DSK, Malik A, Datta A, Mondal H. Radiologic Decision-Making for Imaging in Pulmonary Embolism: Accuracy and Reliability of Large Language Models—Bing, Claude, ChatGPT, and Perplexity. Indian Journal of Radiology and Imaging. 2024;34(4):653-660.
  • Sarangi PK, Datta S, Mondal H. Comment on: ChatGPT: Chasing the Storm in Radiology Training and Education. Indian Journal of Radiology and Imaging. 2024;34(4):792-794.
  • Ellappan K, Datta S, Muthuraj M, Lakshminarayanan S, Pleskunas JA, Horsburgh CR Jr, Salgame P, Hochberg N, Sarkar S, Ellner JJ, Roy G, Jose M, Kumar SV, Joseph NM. Evaluation of Factors Influencing Mycobacterium tuberculosis Complex Recovery and Contamination Rates in MGIT960. Indian Journal of Tuberculosis. 2020;67(4):466-471.

Refereed conference proceedings and book chapters

  • Pate S, Farooq A, Datta S, Sheikh MA, Kumar A, Mishra D. Fine-Grained Rib Fracture Diagnosis with Hyperbolic Embeddings: A Detailed Annotation Framework and Multi-Label Classification Model. In: Medical Image Computing and Computer Assisted Intervention—MICCAI 2025. Lecture Notes in Computer Science, vol. 15974. Springer; 2025:218-227.
  • Dabass M, Chandalia A, Datta S, Mahapatra D. An Attention Learning-Enabled 3D Conditional Generative Adversarial Network for Lung Nodule Segmentation. In: Proceedings of the International Joint Conference on Advances in Computational Intelligence (IJCACI 2022). Algorithms for Intelligent Systems. Springer; 2024:299-312.
  • Dabass M, Chandalia A, Datta S, Mahapatra D. ALE-GAN: A 3D Conditional Generative Adversarial Network with Attention Learning Modules for Lung Nodule Segmentation. In: Advances in Data-Driven Computing and Intelligent Systems. Lecture Notes in Networks and Systems, vol. 890. Springer; 2024:321-332.
  • Dabass M, Chandalia A, Senasi R, Datta S. Attention and Residual-Atrous Convolutional Learning-Based CNN Architecture for Lung Nodule Segmentation and Classification. In: Advances in Data-Driven Computing and Intelligent Systems. Springer; 2024:103-117.
    • DOI: https://doi.org/10.1007/978-981-99-9518-9_8
  • Datta S. AIM in Interventional Radiology. In: Lidströmer N, Ashrafian H, eds. Artificial Intelligence in Medicine. Springer; 2022:459-472.

Preprints and technical reports

  • Datta S*, Buchireddygari D*, Kaza LVC, Bhalke M, Singh K, Pandey A, et al. Radiology & Last Exam (RadLE): Benchmarking Frontier Multimodal AI Against Human Experts and a Taxonomy of Visual Reasoning Errors in Radiology. arXiv preprint. 2025; arXiv:2509.25559. *Equal contribution.
  • Datta S, Buchireddygari D, Bhatti HBS, and the CRASH Lab team. Radiology's Last Exam 2.0: Are We Ready for Autonomous AI in Radiology? CRASH Lab technical report and leaderboard. Released July 2026.

Academic podcasts

  • Dogra S, Datta S. The Full-Time Researcher—Career Pivots, an AJR Podcast Series (Episode 6). AJR Podcasts. Released 24 December 2025.
Sticky Button