Biography
Yu Han is a DPhil candidate in Engineering Science at the University of Oxford, supervised by Prof. Jeroen Bergmann and Prof. Paulo Savaget. Her research sits at the intersection of Artificial Intelligence (AI), Data Science, Natural Language Processing (NLP), Machine Learning, and Regulatory Affairs, with a particular focus on AI-driven solutions for medical regulations. With a multidisciplinary background in Regulatory Affairs (MS, Northeastern University) and Clinical Pharmacy (BS, Capital Medical University), Yu combines expertise in healthcare compliance with advanced AI methodologies. Her work includes developing AI-powered medical device classification systems, qualitative research in regulatory affairs, data-driven analysis of global regulatory frameworks, and multi-agent regulatory AI models,.
Yu has published extensively in leading journals and conferences. Before pursuing her doctorate, she served as a Regulatory Affairs Team Lead at Johnson & Johnson (USA) and held key roles at several U.S. regulatory consulting firms. She played a pivotal role in shaping global registration strategies and led the drafting and submission of regulatory filings, including 510(k), PMA, IND, NDA, and BLA applications. In addition to her research, she is an active peer reviewer and contributes to regulatory and medical science domains, advancing AI applications in healthcare and compliance.
Research Interests
- Generative AI and Large Language Models (LLMs)
- Regulatory Affairs AI/NLP
- Applications Explainable AI (XAI)
- Global Regulatory Frameworks
- AI in Healthcare Innovation
Research Projects
- Regulatory AI Multi-Agent Modelling: Developing a feedback-driven multi-agent system using Large Language Models (LLMs) to simulate and optimise regulatory compliance strategies between manufacturers and authorities.
- AI-Based Medical Device LLM Classification: Building machine learning models to classify medical device software under FDA and NMPA regulatory frameworks, comparing traditional algorithms with state-of-the-art LLMs.
- Regulatory Frameworks for AI-Enabled Medical Device Software: Analysing and comparing global regulatory policies for AI-enabled medical devices to identify trends and opportunities for manufacturers.
- Data-Driven Analysis of AI in Medical Device Software: Investigating large regulatory datasets to identify AI adoption trends in medical device software and visualize market insights.
- Vectorizing Medical Regulations Across Jurisdictions: Using natural language processing (NLP) techniques to quantify and compare regulatory changes across the USA, EU, and China over the past decade.
- Readability Metrics and Regulatory Complexity: Evaluating the complexity of regulatory documents using NLP tools to improve transparency and accessibility for stakeholders.
- Regulatory Affairs Complexity and AI Integration: Designing AI-driven frameworks to address regulatory complexities in medical device development and compliance processes.
