My research lies at the intersection of artificial intelligence, natural language processing (NLP), and biomedical informatics, with a focus on making complex medical data easier to understand and use. In particular, I work with electronic health records (EHRs), which contain valuable but often unstructured clinical information. By developing advanced machine learning and large language model (LLM) techniques, I aim to automatically extract meaningful insights from clinical text, helping healthcare professionals and researchers access important information more efficiently. My work also involves building and organizing medical knowledge systems, known as ontologies, that connect related concepts and improve how medical data is interpreted.

Across my projects, I have designed intelligent systems for identifying medical terms in large-scale datasets, linking them to standardized biomedical concepts, and improving the speed and accuracy of clinical data annotation. For example, I have developed hybrid AI models that combine rule-based methods with modern deep learning approaches to recognize diseases, symptoms, and treatments from real-world clinical notes. I have also explored how large language models can predict relationships within medical knowledge structures, achieving high accuracy and demonstrating their potential in healthcare applications. More broadly, my research aims to bridge the gap between cutting-edge AI technologies and real-world healthcare needs, ultimately supporting better data-driven decision-making and improving patient care. 

See Shuxin Zhou. CV