Qian NIU

Researcher
Matsuo–Iwasawa Laboratory, The University of Tokyo

Trustworthy medical AI: social acceptance, safety and capability evaluation, and AI-driven drug discovery.

Portrait of Qian NIU

About

I am a Researcher at the Matsuo–Iwasawa Laboratory, The University of Tokyo, working on trustworthy medical AI and, increasingly, on AI-driven drug discovery. My work spans how these systems are received in practice and how the models behind them are built and evaluated, published as first or corresponding author in Q1 journals and at top conferences.

My doctoral research in Human Health Sciences at Kyoto University, completed in 2024 and built on a clinical background, took up vaccine hesitancy and vaccination behaviour in Japan. It was supported by a Google PhD Fellowship (Health track; 61 recipients worldwide, 2 in Japan) and a JST SPRING Fellowship, and received the Kyoto University President's Award (Academic Division; 2 university-wide recipients).

Contact & Openings

Collaborations on any of the directions below are always welcome. Write to qian.niu@weblab.t.u-tokyo.ac.jp.

I am also looking for research assistants and interns. Please attach your CV and a short note on what you would like to work on.

Research Agenda

My research runs along three directions: whether society accepts and trusts medical AI (I), what those models are safe and capable enough to do (II), and how far the same approach carries once the language is biological rather than natural (III).

Live the questions now.

— Rainer Maria Rilke, Letters to a Young Poet, 1903

  1. I. Social Acceptance of Medical AI

    My doctoral research pioneered infodemiology studies that mine social-media and search-engine data to address HPV-vaccine hesitancy and COVID-19 vaccination in Japan, via LLM-based stance and sentiment analysis and time-series modelling, culminating in a RAG-based vaccine-communication agent.

  2. II. Safety & Capability Evaluation of Medical LLMs

    In the NEDO-supported Japanese Medical LLM project at the Matsuo–Iwasawa Lab, I lead the development of safety evaluation and build three Japanese clinical benchmarks: JMedEthicBench, JMed48k, and KokushiMD-10. I am also closely involved in the project's data processing and pre-training.

  3. III. AI-Driven Drug Discovery

    This direction extends generative AI from natural language to biological language (genes, proteins, cells), spanning single-cell perturbation response, antibody design, and de novo peptide sequencing, with precision medicine as the target.

Selected Publications

Full list on Google Scholar. * corresponding author  ·  † project lead

  1. Xun Y, Liu J, Niu Q*, Wang X, Yuan Z, Li Z, Zhang Z, Zhao B, Wang S, Li I, Hatakeyama-Sato K, Iwasawa Y, Matsuo Y. JMed48k: A Multi-Profession Japanese Medical Licensing Benchmark for Vision-Language Model Evaluation. under review
  2. Du Y, Niu Q, Yang C, Yu N, Wang S. LIPNovo+: Self-Reflective Latent Imputation for Robust De Novo Peptide Sequencing. under review
  3. Liu J, Li Z, Niu Q*, Zhang Z, Xun Y, Hou W, Wang S, Iwasawa Y, Matsuo Y, Hatakeyama-Sato K. JMedEthicBench: A Multi-Turn Adversarial Benchmark for Japanese Medical Ethics Alignment in LLMs. COLM 2026.
  4. Yang G, Du Y, Hou W, Niu Q*, Wang S*. PRA-PoE: Robust Multimodal Alzheimer's Disease Classification under Arbitrary Modality Missingness. MICCAI 2026.
  5. Yang G, Ding T, Hou W, Xun Y, Du Y, Niu Q*, Wang S*. BrainAnytime: Anatomy-Aware Cross-Modal Pretraining for Brain Image Analysis with Arbitrary Modality Availability. MICCAI 2026.
  6. Zhou Z, Lu Y, Zhu Y, Zhao Y, Niu Q, He B, He L, Yu W, Iwasawa Y. LLM-Guided Adaptive Compensator: Bringing Adaptivity to Robotic Feedback Control with Large Language Model. IEEE Transactions on Automation Science and Engineering, 2026;23:10762-10774.
  7. Wang R, Huang W, Song S, Zhang H, Niu Q, Iwasawa Y, Matsuo Y, Guo J. Beyond In-Distribution Success: Scaling Curves of CoT Granularity for Language Model Generalization. CPAL 2026.
  8. Liu J, Yang S, Ma D, Niu Q†, Nagai-Tanima M, Aoyama T. Japanese AI Agent System on Human Papillomavirus Vaccination: System Design. JMIR Infodemiology, 2026.
  9. Jiang Y, Jin Y, Sun T, Guo H, Zhang Z, Li Y, Niu Q. Enhancing Dermoscopic Image Generation via Multi-Modal Conditional Diffusion Models. Neurocomputing, 2026.
  10. Niu Q, Liu J, Nagai-Tanima M, Aoyama T. Understanding Human Papillomavirus Vaccination Hesitancy in Japan Using Social Media: Content Analysis. Journal of Medical Internet Research, 2025;27:e68881.
  11. Liu J, Yan K, Wang T, Niu Q†, Nagai-Tanima M, Aoyama T. KokushiMD-10: Benchmark for Evaluating Large Language Models on Ten Japanese National Healthcare Licensing Examinations. Clinical MLLMs Workshop, MICCAI 2025.
  12. Niu Q, Liu J, Zhao Z, Onishi M, Kawaguchi A, Bandara A, Harada K, Aoyama T, Nagai-Tanima M. Explanation of Hand, Foot, and Mouth Disease Cases in Japan Using Google Trends Before and During the COVID-19: Infodemiology Study. BMC Infectious Diseases, 2022;22(1):806.
  13. Niu Q, Liu J, Kato M, Nagai-Tanima M, Aoyama T. The Effect of Fear of Infection and Sufficient Vaccine Reservation Information on Rapid COVID-19 Vaccination in Japan: Evidence from a Retrospective Twitter Analysis. Journal of Medical Internet Research, 2022;24(6):e37466.
  14. Niu Q, Liu J, Kato M, Shinohara Y, Matsumura N, Aoyama T, Nagai-Tanima M. Public Opinion and Sentiment Before and at the Beginning of COVID-19 Vaccinations in Japan: Twitter Analysis. JMIR Infodemiology, 2022;2(1):e32335.
  15. Niu Q, Nagata T, Fukutani N, Tezuka M, Shimoura K, Nagai-Tanima M, Aoyama T. Health Effects of Immediate Telework Introduction during the COVID-19 Era in Japan: A Cross-Sectional Study. PLOS ONE, 2021;16(10):e0256530.

Experience & Education

Professional Experience

Education

Languages Chinese (native)  ·  Japanese (JLPT N1)  ·  English (TOEIC 860)

Awards & Fellowships

Academic Service

Journal reviewer

  • IEEE Journal of Biomedical and Health Informatics
  • Journal of Medical Internet Research
  • JMIR Public Health and Surveillance
  • JMIR Infodemiology
  • JMIR Cancer
  • Frontiers in Public Health

Conference reviewer

  • ACL (ARR)
  • MICCAI