EmoPatient
An emotion-directed patient simulator that generates dynamic, interpretable responses for realistic palliative care communication training.
Patient simulation · Multi-agent systems · Emotional modeling
Master's student at UT Austin
I work on reliable AI for health.
I am a master's student at The University of Texas at Austin. I conduct research at the AI Health Lab with Prof. Ying Ding, and at Dell Medical School's Department of Internal Medicine with Dr. W. Michael Brode.
In summer 2026, I was a research intern in Mayo Clinic's Department of Artificial Intelligence and Informatics. My work develops multimodal, efficient, and rigorously evaluated AI systems for complex biomedical data and clinical decision support.
I am currently seeking Ph.D. opportunities.
I study how AI systems can learn from complex clinical data while remaining useful, interpretable, and trustworthy in real care settings.
My work spans longitudinal EHR modeling, clinical NLP, multimodal learning, retrieval-augmented generation, and multi-agent systems. I pair model development with careful ablation, error analysis, and clinician-in-the-loop evaluation.
Before joining UT Austin, I earned a Bachelor of Engineering in Vehicle Engineering from Tongji University. That training continues to shape the systems-oriented way I approach biomedical AI research.
Current work across patient simulation, evidence-grounded clinical support, target trial emulation, and longitudinal risk modeling.
An emotion-directed patient simulator that generates dynamic, interpretable responses for realistic palliative care communication training.
Patient simulation · Multi-agent systems · Emotional modeling
A clinical support system that brings together curated consensus guidance, PubMed literature, clinical trials, and living systematic-review evidence.
Retrieval-augmented generation · Clinical NLP · Evidence provenance
A contract-governed agent framework that translates target trial protocols into site-specific EHR study definitions with staged skills, validation, and clinician review.
Clinical agents · Target trial emulation · Human-in-the-loop review
An extractor-scorer pipeline that turns three years of multimodal EHR timelines into clinically relevant events for five-year cardiovascular risk prediction.
Longitudinal EHR · Risk modeling · Efficient adaptation
AMIA 2026
Yining Wu, Tianshu Du, Jinrui Fang, Chi Zhang, Sonal Admane, Ying Ding
DAIH @ COLM 2026
Yining Wu, Philip DiGiacomo, Ying Ding, William Brode
Preprint · 2026
Xu Yang, Zhizhou Sha, Junbo Li, Jian Yu, Yifan Sun, Matthew Zhao, Jinrui Fang, Xinyue Guo, Yining Wu, Xu Hu, Yifu Luo, Qiang Liu, Zhangyang Wang
May–Aug 2026
Mayo Clinic · Jacksonville, Florida
Clinical risk modeling and automated EHR-based target trial emulation.
Jan 2026–present
AI Health Lab, UT Austin
Emotion-directed patient simulation for palliative care communication training.
Oct 2025–Jun 2026
Dell Medical School, UT Austin
Evidence-grounded clinical decision support for Long COVID.
2025–2027
The University of Texas at Austin
GPA: 4.0 / 4.0
2020–2025
Tongji University
GPA: 90.5 / 100
Python · SQL · PyTorch · Transformers · Clinical NLP · Multimodal learning · Multi-agent systems · Retrieval-augmented generation · Ablation and error analysis
I welcome conversations about biomedical AI research, clinical collaborations, and Ph.D. opportunities.