Sunyang Fu, PhD, MHI


Assistant Professor & Associate Director of Team Science, Center for Translational AI Excellence and Applications in Medicine (TEAM-AI), McWilliams School of Biomedical Informatics, UTHealth Houston

Google Scholar Scholar Twitter Twitter LinkedIn Linkedin GitHub Github

Email: Sunyang.Fu@uth.tmc.edu

Recent News

Sep 13 2024: We received two pilot awards from the Artificial Intelligence and Technology Collaboratories (AITC) for Aging Research program and UTHealth Insitution of Aging to study acute heart failure exacerbation and adverse drug events among Older Adults.

Aug 9 2024: Our study A Generalist Vision-Language Foundation Model for Diverse Biomedical Tasks has been accepted by Nature Medicine.

June 20 2024: Our study error taxonomy was featured by JAMIA and selected as AHRQ collection.

Mar 18 2024: Honored to receive NIDUS II LOI Award to improve the differential detection between delirium and dementia

Mar 17 2024: Honored to receive AMIA IS 24 SPC Award.

Sep 1 2023: Excited to be selected as a Leadership Fellow by the NIH AIM-AHEAD initiative.

Jul 1 2023: Very excited to join UTHealth McWilliams School of Biomedical Informatics as a faculty member.

Mar 18 2023: Our CTS study is featured by the American Society for Clinical Pharmacology and Therapeutics and Clinical and Translational Science, as well as Mayo Clinic's Research Magazine.


Bio

I am an Assistant Professor and Associate Director of Team Science at the Center for Translational AI Excellence and Applications in Medicine (TEAM-AI), McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston. I am also affiliated with UTHealth Institute on Aging, Network for Investigation of Delirium: Unifying Scientists, and Mayo Clinic, Division of Epidemiology. Additionally, I was selected as the 2023-24 Leadership Fellow with the National Institute of Health’s Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity (AIM-AHEAD). The overarching goal of my research is to accelerate, improve and govern the secondary use of Electronic Health Records (EHRs) for clinical and translational research toward high throughput, reproducible, fair, and trustworthy discoveries. I have significant collaborative research experience in aging, cancer, and musculoskeletal diseases and procedures. Previously, I was a Sr. Data Science Analyst at the Department of AI and Informatics, Mayo Clinic. I obtained my Ph.D. at the University of Minnesota, M.H.I. at the University of Michigan, and B.B.A. at the University of Iowa.

Prospective Students

I am looking for talented graduate students who are interested in:

Feel free to email me with information about your relevant experience and goals. For more information, please visit https://sbmi.uth.edu/prospective-students/

Research Highlights


Improving Detection of Aging-related Outcomes Using Real-World Data

Overview: Many geriatric syndromes and aging-related outcomes, such as delirium, falls, frailty, and cognitive impairment, are underdiagnosed in clinical practice and are not routinely coded for billing. We developed hybrid methods to systematically extract and standardize these outcomes from real-world EHR data, supporting both clinical research and quality improvement initiatives. Our ultimate goal is to advance Age-Friendly Health Systems through innovative digital solutions.

Related Publications:

Fu S, Jia H, Vassilaki M, Keloth VK, Dang Y, Zhou Y, Garg M, Petersen RC, St Sauver J, Moon S, Wang L. FedFSA: Hybrid and federated framework for functional status ascertainment across institutions. Journal of Biomedical Informatics. 2024 Apr 1;152:104623. https://doi.org/10.1016/j.jbi.2024.104623

Fu S, Lopes GS, Pagali SR, Thorsteinsdottir B, LeBrasseur NK, Wen A, Liu H, Rocca WA, Olson JE, St. Sauver J, Sohn S. Ascertainment of delirium status using natural language processing from electronic health records. The Journals of Gerontology: Series A. 2022 Mar 1;77(3):524-30. https://doi.org/10.1093/gerona/glaa275

Fu S, Thorsteinsdottir B, Zhang X, Lopes GS, Pagali SR, LeBrasseur NK, Wen A, Liu H, Rocca WA, Olson JE, Sauver JS. A hybrid model to identify fall occurrence from electronic health records. International journal of medical informatics. 2022 Jun 1;162:104736. https://doi.org/10.1016/j.ijmedinf.2022.104736


Accelerate the Translation of Natural Language Processing for Clinical Research

Overview: Manually reviewing patient records for extracting information for clinical research is labor-intensive and requires specialized knowledge possessed by highly trained medical professionals. Natural language processing (NLP) tools are distinctive in their ability to extract critical information from raw texts in EHR. My research focuses on developing methodologies to improve the validity, implementability, and portability of clinical NLP applications.

Related Publications:

Fu S, Wang L, He H, Wen A, Zong N, Kumari A, Liu F, Zhou S, Zhang R, Li C, Wang Y. A taxonomy for advancing systematic error analysis in multi-site electronic health record-based clinical concept extraction. Journal of the American Medical Informatics Association. 2024 May 14:ocae101. https://doi.org/10.1093/jamia/ocae101

Fu S, Wang L, Moon S, Zong N, He H, Pejaver V, Relevo R, Walden A, Haendel M, Chute CG, Liu H. Recommended practices and ethical considerations for natural language processing‐assisted observational research: A scoping review. Clinical and translational science. 2023 Mar;16(3):398-411. https://doi.org/10.1111/cts.13463

Fu S, Leung LY, Raulli AO, Kallmes DF, Kinsman KA, Nelson KB, Clark MS, Luetmer PH, Kingsbury PR, Kent DM, Liu H. Assessment of the impact of EHR heterogeneity for clinical research through a case study of silent brain infarction. BMC medical informatics and decision making. 2020 Dec;20:1-2. https://doi.org/10.1186/s12911-020-1072-9


Enhancing Multi-Institutional EHR Data Management and Use for Reproducible and Valid Discoveries

Overview: Recent advancement in healthcare AI identifies the need for detailed provenance information of data obtained for training and validating AI models. Therefore, secondary use of EHR for clinical research leveraging AI technologies requires the documentation of the provenance information that captures the process of the retrieval and organization of raw data as well as the extraction and annotation of training data. To enable and accelerate EHR-based clinical research through clinical NLP, I have promoted translational efforts that integrate informatics-driven approaches and multi-institutional collaboration, employing multi-phased methods and a people-centric approach to develop processes, standards, and best practices that ensure reproducible and valid research.

Related Publications:

Fu S, Wen A, Schaeferle GM, Wilson PM, Demuth G, Ruan X, Liu S, Storlie C, Liu H. Assessment of data quality variability across two EHR systems through a case study of post-surgical complications. AMIA Summits on Translational Science Proceedings. 2022;2022:196. PMC9285181

Fu S, Vassilaki M, Ibrahim OA, Petersen RC, Pagali S, St Sauver J, Moon S, Wang L, Fan JW, Liu H, Sohn S. Quality assessment of functional status documentation in EHRs across different healthcare institutions. Frontiers in Digital Health. 2022 Sep 27;4:958539. https://doi.org/10.3389/fdgth.2022.958539

Fu S, Wen A, Pagali S, Zong N, St Sauver J, Sohn S, Fan J, Liu H. The implication of latent information quality to the reproducibility of secondary use of electronic health records. InMEDINFO 2021: One World, One Health–Global Partnership for Digital Innovation 2022 (pp. 173-177). IOS Press. 10.3233/SHTI220055

Current Students

  • Taylor Harrison, MBA, (PhD student, University of Minnesota, Co-advised by Dr. Hongfang Liu)
  • Ethan Zhang, (Undergraduate, Northwestern University)