Chih-Lin Chi, PhD, MBA

Research
Health Data Science
Chih-Lin Chi
[email protected]
612-624-5113

8-102 Phillips-Wangensteen Building
516 Delaware St. SE
Minneapolis, MN 55455
United States

I am a biomedical informatician specializing in the creation and implementation of approaches, methods, and software tools designed to discover information and evidence from diverse medical data including electronic medical records, hospital discharge notes, longitudinal cohort studies, laboratory tests, and genetic data. I am particularly interested in understanding how a person's individual characteristics influence the outcome of multiple medical treatments; and how an individual, care team, and hospital network can select and implement the treatment decision that will maximize the overall healthcare outcome. My background in machine learning, operations research, artificial intelligence, and statistics has prepared me to create sophisticated computational approaches to discover evidence and optimize the results of this type of personalized health management from high-dimensional medical data including personalized prevention, diagnosis, treatment, and prognosis.

To realize the personalized health management in clinical settings, the highly complex mathematical functions used to accurately model complicated medical events will be converted to decision support rules. These rules indicate which treatment option most improves outcomes for a particular type of individuals. The transparent property of rules further allows clinical validation by domain experts, in-depth clinical studies, and clinical trials. My recent efforts also include using clinical trial simulations and genetic study to gain insight into such computational evidence and understand why a particular type of patients has the optimal outcome when receiving a certain treatment option.

My agenda of the personalized health management studies includes four elements that complement each other: (1) developing translational research platform for the personalized health management starting from evidence discovery from medical data to clinical validation and implementation, (2) including omics data study aiming to strengthen such personalized health management evidence, (3) improving computational methods to incorporate realistic factors (such as factors that have been discussed in the past and ongoing projects: costs, compliance, disease progression, distance to clinics, and other clinical limitations) to support practical settings, and (4) applying abovementioned frameworks and approaches to multiple-center studies to improve robustness of the evidence.

Instead of inventing new treatment that typically takes millions (if not, billions) of dollars and years of efforts, the personalized health management seeks to identify improved-outcome evidence from medical data and apply the evidence to support personalized care.

Education

Associate Professor, Population Health and Systems Cooperative

Core Faculty, Institute for Health Informatics (IHI)

Postdoctoral Fellowship, Harvard Medical School (Biomedical Informatics, Laboratory for Personalized Medicine, Center for Biomedical Informatics), 2013

PhD, University of Iowa (Machine Learning in Healthcare: Health Informatics), 2009

MBA, Feng Chia University (Marketing and Management Information Systems), 2000

BS, National Chung-Hsing University (Zoology), 1998

Expand all

Research

Research Summary/Interests

  • Personalized health management
  • Knowledge discovery and data mining
  • Machine learning and operations research
  • Modeling and simulation
  • Translational research
  • Decision-support systems
  • Metabolomic data analysis

 

Research Funding Grants

Selected Grants:

Anticlotting Simulations to predict optimal treatment for sub-populations
Role: Co-Principal Investigator
Funding Agency: NIH/NLM (R01 1R01LM011566-01)
Project Dates: 09/15/2013–09/14/2017

Discover Evidence of Personalized Alzheimer’s Treatment from Electronic Health Records
Role: Principal Investigator
Funding Agency: University of Minnesota Grant-in-Aid
Project Dates: 1/1/2015 – 6/30/2016

Discover Evidence of Personalized Prevention of Statin Intolerance from Electronic Health Records
Role: Principal Investigator
Funding Agency: University of Minnesota Academic Health Center Seed Grants
Project Dates: 1/1/2015 – 12/31/2015

Feasibility of examining hidden patterns in Omaha System data: Original research within a multi-user data collaborative
Role: Principle Investigator
Funding Agency: University of Minnesota Foundation through the Omaha System Partnership
Project Dates:07/1/14-12/31/15

Publications

Selected Publications:

  • Chi, C.-L., Oh, W., Borson, S. Feasibility study of a machine learning approach to predict dementia progression. Proceedings of the IEEE International Conference on Healthcare Informatics (ICHI), page 450, Dallas, TX, October 2015.
  • Monsen, K.A., Peterson, J.J., Mathiason, M.A., Kim, E., Lee, S., Chi, C.-L., Pieczkiewicz, D.S. Data visualization techniques to showcase nursing care quality. Computers Informatics Nursing, 33(10):417-26, October 2015
  • Rosenblum Lichtenstein, J.H., Hsu, Y.H., Gavin, I.M., Donaghey, T.C., Molina, R.M., Thompson, K.J., Chi, C.-L, Gillis, B.S., Brain, J.D. Environmental mold and mycotoxin exposures elicit specific cytokine and chemokine responses. PLoS One, 10(5), May 2015.
  • Chi, C.-L., & Tonellato, P.J. Discover Improved-Outcome Evidence for Personalized Treatment from Electronic Health Records, Proceedings of the 2014 American Medical Informatics Association Annual Symposium, page 1344, Washington, DC, November 2014.
  • Chi, C.-L., & Tonellato, P.J. Optimize Warfarin Treatment through Prediction and Simulation. Proceedings of the 8th INFORMS Workshop on Data Mining and Health Informatic, Minneapolis, MN, October 2013.
  • Chi, C.-L., & Tonellato, P.J. Optimize Warfarin Treatment by Tailoring Protocol. Proceedings of AMIA Summit on Clinical Research Informatics, San Francisco, CA, March 2013.
  • Fusaro, V.A., Patil, P., Chi, C.-L., Contant, C.F., & Tonellato, P.J. A systems approach to designing effective clinical trials using simulations, Circulation, 127(4):517-526, January 2013.
  • Chi, C.-L., Street, W.N., & Robinson, J.G. Individualized patient-centered lifestyle recommendations: An expert system for communicating patient specific cardiovascular risk information and prioritizing lifestyle options. Journal of Biomedical Informatics, 45(6):1164-1174, December 2012.
  • Chi, C.-L., Fusaro, V.A., Patil, P., Crawford M. A., Contant, C.F., & Tonellato, P.J. A simulation platform to examine heterogeneity influence on treatment. Proceedings of AMIA Summit on Translational Science, pages 19-24, San Francisco, CA, March 2012.
  • Chi, C.-L., Fusaro, V.A., Patil, P., Crawford, M.A., Contant, C.F., & Tonellato, P.J. An approach to optimal individualized warfarin treatment through clinical trial simulations. Proceedings of the 5th IEEE Cairo International Biomedical Engineering Conference (CIBEC 10), page 16-18, Cairo, Egypt, December 2010.
  • Chi, C.-L., Patil, P., Fusaro, V.A., Kos, P.J., Pivovarov, R., Contant, C.F., & Tonellato, P.J. Simulated Comparison of Warfarin Treatment Protocols. Proceedings of the 2010 American Medical Informatics Association Annual Symposium, page 1009, Washington, DC, November 2010.
  • Chi, C.-L., Kos, P.J., Fusaro, V.A., Pivovarov, R., Patil, P., & Tonellato, P.J. Mining personalized medicine algorithms with surrogate algorithm tags. Proceedings of the First ACM International Health Informatics Symposium, pages 474-478, Arlington, VA, November 2010.
  • Chi, C.-L., Street, W.N., & Katz, D.A. A decision support system for cost-effective diagnosis. Artificial Intelligence in Medicine, 50(3):149-161, November 2010.
  • Chi, C.-L. & Street, W.N. The optimal diagnostic decision sequence. American Medical Informatics Association Annual Symposium, page 902, Washington, DC, November 2008.
  • Chi, C.-L., Street, W.N., & Ward, M.M. Building a hospital referral expert system with a prediction and optimization-based decision support system algorithm. Journal of Biomedical Informatics 41(2):371-386, April 2008.
  • Chi, C.-L., Street, W.N., & Wolberg, W.H. Application of articial neural network-based survival analysis on two breast cancer datasets. Proceedings of the 2007 American Medical Informatics Association Annual Symposium, pages 130-134, Chicago, IL, November 2007.
  • Chi, C.-L. & Street, W.N. A data mining technique for risk-stratification diagnosis. Proceedings of the 2007 American Medical Informatics Association Annual Symposium, page 909, Chicago, IL, November 2007.