Dr. Su' research, lying in the intersection between data science and medicine, aims at developing novel computational approaches (particularly machine learning [ML] and deep learning [DL]) to derive new insights from diverse, multi-modal health data, such as real-world Electronic Health Records (EHRs), multi-omics data, medical imaging, and social determinants of health (SDoH), for accelerating human disease research and health care.
One of the major research topics I am working on is the development of ML&DL-based data-driven approaches for disentangling disease complexity and identifying tailored treatments towards precision medicine. I have developed advanced ML&DL approaches for integrative analyses of multimodal biomedical data from various data sources. In addition, due to the potential challenge on working with biomedical data (e.g., uncertainty, sparsity, heterogeneity, etc.), the insights derived from pure data-driven analysis may not be reliable enough. Therefore, I have also devoted significant efforts on collecting and summarizing biomedical knowledge from various sources (e.g., literature, experiments and existing biomedical knowledge bases) and building a comprehensive biomedical knowledge graph (BKG), which can effectively compensate and enhance the data-driven insights.
My approaches have been applied to studies of complex human diseases or health conditions, including COVID-19, neurodegenerative diseases like Parkinson’s disease (PD) and Alzheimer’s diseases (AD), mental illness & suicide, skin inflammatory disease like Hidradenitis Suppurativa and other fibrotic skin disorders, inflammatory bowel disease, sepsis, etc.
Biography
Dr. Su is an Assistant Professor of Health Informatics and Artificial Intelligence in the Department of Population Health Sciences. His research focuses on developing machine learning and AI methods to derive novel insights from multimodal biomedical data. He also conducts research on biomedical knowledge graph (BKG) construction and BKG-based learning approaches to advance biomedical data science. He has published in leading medical journals, including JAMA Pediatrics, npj Digital Medicine, Cancer Cell, and Immunity, as well as in major AI venues such as NeurIPS. Dr. Su's research has been supported by the NIH, MJFF, dkNET, and the Department of Population Health Sciences.
