Publications

Journal Articles & Preprints

Trustworthiness Evaluation of Medical Vision-Language Models: A Scoping Review of Robustness, Grounding, Hallucination, and Uncertainty

Published in JMIR AI preprint; manuscript under review, 2026

A PRISMA-ScR review of how medical VLM research evaluates robustness, visual grounding, hallucination, and uncertainty—and where safety evidence remains thin.

Recommended citation: Sadanandan, B., Karimi, A., Upadhayay, B., & Behzadan, V. (2026). Trustworthiness evaluation of medical vision-language models: A scoping review of robustness, grounding, hallucination, and uncertainty. Manuscript under review. doi:10.2196/preprints.102330.
Download Paper

Multimodal Deep Learning for Early Prediction of Patient Deterioration in the ICU

Published in Journal of Data Science, 2025

Multimodal deep learning models that combine clinical signals to predict ICU patient deterioration before it becomes critical.

Recommended citation: Multimodal Deep Learning for Early Prediction of Patient Deterioration in the ICU. Journal of Data Science, 2025.

Promise of Data-Driven Modeling and Decision Support for Precision Oncology and Theranostics

Published in arXiv preprint, 2025

Integration of data-driven PBPK modeling with reinforcement learning for dynamic treatment optimization in cancer therapy.

Recommended citation: Sadanandan, B., & Behzadan, V. (2025). Promise of Data-Driven Modeling and Decision Support for Precision Oncology and Theranostics. arXiv preprint arXiv:2505.09899.
Download Paper

Machine Learning for Anastomotic Leak Prediction: A Systematic Review and Experimental Validation

Published in International Journal of Trend in Scientific Research and Development 8(3), 12, 2024

A systematic review and experimental validation of machine learning techniques for predicting anastomotic leak in surgical procedures.

Recommended citation: Sadanandan, B. (2024). Machine Learning for Anastomotic Leak Prediction: A Systematic Review and Experimental Validation. International Journal of Trend in Scientific Research and Development, 8(3), 12.

Conference Papers

Attention Without Grounding: Causal Evaluation of Visual Explanations in Medical VLMs

Published in iMIMIC Workshop, MICCAI 2026, 2026

Attention maps can look convincing while playing no causal role in the answer. We test whether visual explanations in medical VLMs actually reflect what the model uses.

Recommended citation: Sadanandan, B., & Behzadan, V. (2026). Attention Without Grounding: Causal Evaluation of Visual Explanations in Medical VLMs. iMIMIC Workshop, MICCAI 2026.
Download Paper

Mechanistically Guided LoRA Improves Paraphrase Consistency in Medical Vision-Language Models

Published in Conference on Health, Inference, and Learning (CHIL) 2026, 2026

Sparse-autoencoder and residual-stream analyses identify candidate mechanisms for paraphrase flips; a targeted LoRA cuts the flip rate while touching 0.1% of parameters.

Recommended citation: Sadanandan, B., & Behzadan, V. (2026). Mechanistically guided LoRA improves paraphrase consistency in medical vision-language models. Conference on Health, Inference, and Learning (CHIL) 2026. arXiv:2603.00148.
Download Paper

Predictive Entropy as a Joint Screen for Error and Paraphrase Instability in Medical Vision-Language Models

Published in UNSURE Workshop, MICCAI 2026 (poster), 2026

A single predictive-entropy signal can rank both likely errors and likely paraphrase flips, providing a bounded readiness screen for two failure modes.

Recommended citation: Sadanandan, B., & Behzadan, V. (2026). Predictive entropy as a joint screen for error and paraphrase instability in medical vision-language models. UNSURE Workshop, MICCAI 2026 (poster).
Download Paper

When Chain-of-Thought Backfires: Evaluating Prompt Sensitivity in Medical Language Models

Published in 2AI Conference 2026, 2026

Evaluating how chain-of-thought prompting can paradoxically degrade medical language model performance.

Recommended citation: Sadanandan, B., & Behzadan, V. (2026). When chain-of-thought backfires: Evaluating prompt sensitivity in medical language models. 2AI Conference 2026. arXiv:2603.25960.
Download Paper

Consistent but Dangerous: Per-Sample Safety Classification Reveals False Reliability in Medical Vision-Language Models

Published in MedReasoner Workshop, CVPR 2026, 2026

A model can answer consistently while ignoring the medical image. Per-sample safety classification separates real reliability from text shortcuts.

Recommended citation: Sadanandan, B., & Behzadan, V. (2026). Consistent but Dangerous: Per-Sample Safety Classification Reveals False Reliability in Medical Vision-Language Models. MedReasoner Workshop, CVPR 2026.
Download Paper

PSF-Med: A Clinician-Audited Benchmark for Paraphrase Sensitivity in Medical Vision-Language Models

Published in MMFM-BIOMED Workshop, CVPR 2026, 2026

A benchmark of 92,856 evaluation question-paraphrase pairs showing that medical VLMs can flip their diagnosis when a clinically equivalent question is reworded.

Recommended citation: Sadanandan, B., Behzadan, V., Jayan, L., & Kurup, A. G. (2026). PSF-Med: A clinician-audited benchmark for paraphrase sensitivity in medical vision-language models. MMFM-BIOMED Workshop, CVPR 2026. arXiv:2602.21428.
Download Paper

VSF-Med: A Vulnerability Scoring Framework for Medical Vision-Language Models

Published in IEEE ISBI 2026 (pilot abstract), 2026

A vulnerability scoring framework for evaluating the robustness of medical vision-language models against adversarial attacks.

Recommended citation: Sadanandan, B., & Behzadan, V. (2025). VSF-Med: A vulnerability scoring framework for medical vision-language models. IEEE ISBI 2026 (pilot abstract). arXiv:2507.00052.
Download Paper

Comparative Study of Generative Models for Early Detection of Failures in Medical Devices

Published in 2024 8th International Conference on Medical and Health Informatics, 2024

A comparative study of GAN, VAE, and HMM for fault detection in medical devices using data-driven digital twins.

Recommended citation: Sadanandan, B., Arghavani Nobar, B., & Behzadan, V. (2024). Comparative Study of Generative Models for Early Detection of Failures in Medical Devices. 2024 8th International Conference on Medical and Health Informatics.
Download Paper

Patents

System and Method for Machine Learning Method to Verify Consistent Staple Line Delivery

Published in WO Patent WO2025027493A1, 2025

Patent for machine learning-based verification of consistent staple line delivery in surgical devices.

Recommended citation: Kumar, B., Bidarahalli, P.K., & Miesse, A.M. (2025). System and method for machine learning method to verify consistent staple line delivery. WO Patent WO2025027493A1.

Systems and Methods for Machine Learning Control of a Surgical Device

Published in WO Patent WO2023223258A2, 2023

Patent for machine learning-based control systems for surgical devices.

Recommended citation: Miesse, A.M., Sadanandan, B.K., Evans, C.K., Knapp, R.H., & Jalaja, N.L.V. (2023). Systems and methods for machine learning control of a surgical device. WO Patent WO2023223258A2.

Binesh K Sadanandan

Publications

Journal Articles & Preprints

Conference Papers

Patents