About Me

I am a PhD Student at VinUniversity, specializing in smart healthcare research under the College of Engineering & Computer Science. My primary focus is on developing a Unified Framework for Trustworthy and Reliable AI in Medical Diagnostics.

With a strong background in Deep Learning and Computer Vision, I explore AI applications in healthcare and other high-impact fields. Specifically, I leverage expertise in Explainable AI (XAI) and Visual Question Answering (VQA) to address complex, real-world clinical challenges such as digital phenotype monitoring and neurological/mental health disease assessments in low-resource settings.

Beyond research, I have experience leading R&D teams in building interpretable multimodal systems and full-stack enterprise applications.

Research Interests

Smart Healthcare AIExplainable AIResponsible AIVisual Question Answering

🔥News

2026.09
Vinuni_CVMAIL Wins 1st Place Overall at the MICCAI 2026 MedReason Challenge– Vinuni_CVMAIL ranked 1st Place Overall in the MedReason Challenge 2026, while our mixed-format medical VQA paper was accepted to the MedReason Individual Method Paper Track at MICCAI 2026. The organizers state that accepted papers are scheduled for inclusion in the forthcoming Springer LNCS proceedings.
2026.02
Handling Supervision Scarcity Paper at ISBI 2026– Our work on long-tailed and zero-shot chest X-ray classification is listed at ISBI 2026 and addresses supervision scarcity across in-distribution and unseen disease categories.
2025.12
Completed Master’s Studies at VNU University of Engineering and Technology– Completed my full-time master’s studies at VNU University of Engineering and Technology in December 2025.
2025.11
VietMEAgent at FAIR 2025– VietMEAgent is listed at FAIR 2025 and introduces a culturally aware multimodal explanation framework for Vietnamese visual question answering.
2025.11
VinDr-CXR-VQA Preprint Released– We released VinDr-CXR-VQA, a chest X-ray visual question answering dataset with spatial grounding and clinical reasoning explanations, as an arXiv preprint.
2025.10
MedXplain-VQA at ICHST 2025– MedXplain-VQA is listed at ICHST 2025 and introduces a multi-component framework for explainable medical visual question answering.

📝Selected Publications

Candidate-Expanding Routing with Permutation-Stabilized Experts for Mixed-Format Medical VQA

Hai-Dang Nguyen, Huy-Hieu Pham

MedReason Individual Method Paper Track, MICCAI 2026 (accepted; Springer LNCS forthcoming)

Handling Supervision Scarcity in Chest X-ray Classification: Long-Tailed and Zero-Shot Learning

Ha-Hieu Pham, Hai-Dang Nguyen, Thanh-Huy Nguyen, Min Xu, Ulas Bagci, Trung-Nghia Le, Huy-Hieu Pham

IEEE International Symposium on Biomedical Imaging (ISBI 2026)

VietMEAgent: Culturally-Aware Few-Shot Multimodal Explanation for Vietnamese Visual Question Answering

Hai-Dang Nguyen, Minh-Anh Dang, Minh-Tan Le, Minh-Tuan Le

The 18th National Conference on Fundamental and Applied IT Research (FAIR 2025)

MedXplain-VQA: Multi-Component Explainable Medical Visual Question Answering

Hai-Dang Nguyen, Minh-Anh Dang, Minh-Tan Le, Minh-Tuan Le

The 2nd International Conference on Health Science and Technology (ICHST 2025)

VinDr-CXR-VQA: A Visual Question Answering Dataset for Explainable Chest X-Ray Analysis with Multi-Task Learning

Hai-Dang Nguyen, Ha-Hieu Pham, Hao T. Nguyen, Huy-Hieu Pham

arXiv preprint arXiv:2511.00504

Deep Reinforcement Learning vs. Mean-Variance Optimization

Le, T., Nguyen, H. D.

Accepted for presentation at the 9th Vietnam International Conference in Finance (VICIF 2025)


💼Experience

Teaching Assistant

VinUniversity · Full-time

Sep 2025 – Present

Full-time · On-site · Hanoi, Vietnam.

Research Assistant

VinUniversity · Full-time

Sep 2025 – Present

Full-time · On-site · Hanoi, Vietnam.

PhD Student

VinUniversity · Full-time

Sep 2025 – Present

Full-time · On-site · Hanoi, Vietnam.

Researching trustworthy and reliable AI for medical diagnostics, including explainable AI, medical visual question answering, computer vision, LLMs, and AI agents.

Master’s Student

VNU University of Engineering and Technology · Full-time

Sep 2024 – Dec 2025

Full-time · On-site · Hanoi, Vietnam.

Completed the master’s program in December 2025.

Software Application Developer

T&T Group · Full-time

May 2023 – Sep 2025

Full-time · On-site · Hanoi, Vietnam.

Software Application Developer at T&T Group, working with technologies including SQL and .NET Core.


✍️Blog

Oct 23, 2026

Clinical AI evaluation needs an independent comparison

A UCLA study shows why similarity to an edited AI draft differs from independent clinical-message evaluation, and why BE…

Read more

Oct 20, 2026

LiON Liver CT Study Tests AI as an Additional Reader

A 2026 LiON study tested AI-assisted liver CT review in 10,333 patients. Its single-arm design leaves comparative clinic…

Read more

Oct 20, 2026

AlphaGenome Paper Maps DNA Changes to Molecular Effects

The 2026 AlphaGenome paper predicts molecular effects of DNA variation across multiple tasks. Its research benchmarks ar…

Read more

Oct 20, 2026

MASAI Trial Tests AI Mammography Beyond Detection Rates

The 2026 MASAI analysis found non-inferior interval-cancer rates with AI-supported mammography, higher sensitivity and u…

Read more

Oct 20, 2026

AMIE Cardiology Study Improves Expert-Rated Assessments

A 2026 AMIE study found better expert-rated cardiology assessments using retrospective cases, while leaving real-world p…

Read more

Oct 20, 2026

Kenya AI Trial Finds No Significant Patient-Outcome Benefit

A 2026 randomized trial of AI Consult found no significant reduction in 14-day treatment failure. Here is what the evide…

Read more

Oct 20, 2026

Diffusion Controller Research Explains a New Way to Steer Images

Google highlighted Diffusion Controller in September. The March preprint explains its control framework and the access l…

Read more

Oct 20, 2026

Anthropic Expands Evaluator Access as METR Flags Evidence Gaps

Anthropic announced deeper evaluator access, while METR’s October 20 review highlights why scope, publication rights and…

Read more

Oct 20, 2026

Gemini 4 Argon Robotics Demos Put Simulation in the Loop

Google’s October robotics demonstrations connect visual reasoning, simulation and robot control. Here is what researcher…

Read more

Oct 20, 2026

EmbeddingGemma 2 Brings Multimodal Search to Local Devices

EmbeddingGemma 2 combines text, images, audio and video for local retrieval. Its model card highlights practical trade-o…

Read more

Oct 20, 2026

Google Deep Review Opens AI Document Review to More Researchers

Google launched Deep Review on October 14. Here is what its document-checking workflow offers and why researchers still …

Read more

Oct 20, 2026

MedGemma reaches Nature Medicine: strong benchmarks, not a clinical green light

A newly published MedGemma paper reports promising medical-AI benchmark results. Its findings support further developmen…

Read more