Dr. Joyce Lee is an innovative leader in bringing thoughtful design to academic medicine. She believes that using patient-centered participatory design and harnessing the maker movement creates opportunities for innovation.
What drew you to MiCHAMP?
I am a diabetes specialist and health services researcher with a research agenda that focuses on diabetes and childhood obesity. I am interested in understanding how technology systems and artificial intelligence might reshape the way we deliver healthcare in the future. I was drawn to MICHAMP because of the opportunity to interact with a multidisciplinary group of investigators who are applying machine learning and artificial intelligence to a diverse group of clinical problems.
What specific expertise do you bring to MiCHAMP?
I have expertise in the application of human-centered design for healthcare, and I deeply interested in the design of artificial intelligence algorithms. How might algorithms be designed to gain the trust and confidence of providers? How might they be integrated into the workflow to support rather than “replace” healthcare providers? How might the Electronic Health Record be used as a tool to support the adoption and utilization of AI in clinical delivery systems? These are questions that the process of design can address to help realize the promise of AI. I am also excited about my new role as ACMIO for Pediatric Research. Clinical data collected in the EHR as part of regular care needs to feed into AI algorithms, so it’s critical to think about an optimally designed system for clinical delivery/operations and research, and this definitely involves wrangling with the design of Epic.
What are your research interests and how do they tie into MiCHAMP?
My area of expertise is in diabetes. Diabetes is the disease that is most poised to benefit from a computing revolution, given the continuous stream of data collected by personal diabetes devices, yet this data is practically ignored by patients/caregivers and clinicians except for 3-4 times a year at the diabetes clinic visit where the data is viewed and stored as a pdf. I am therefore collaborating with machine learning/artificial intelligence experts (including Jenna Wiens) who are developing algorithms that use data from the EHR, insulin pump, and continuous glucose monitors to assist providers with automated recommendations for insulin dose changes for patients between clinic visits.
Gavrila, V., Garrity, A., Hirschfeld, E., Edwards, B. and Lee, J. (2019). Peer Support Through a Diabetes Social Media Community. Journal of Diabetes Science and Technology, 13(3), pp.493-497. (PubMed)
Rogers, M., Kim, C., Lee, J., Basu, T. and Tipirneni, R. (2019). Private Insurance Coverage for Diabetes Before and After Enactment of the Preexisting Condition Mandate of the Affordable Care Act, 2005–2016. American Journal of Public Health, 109(4), pp.562-564. (PubMed)
Rogers, M., Kim, C., Tipirneni, R., Basu, T. and Lee, J. (2019). Duration of Insulin Supply in Type 1 Diabetes: Are 90 Days Better or Worse Than 30 Days?. Diabetes Spectrum, 32(2), pp.139-144. (PubMed)
Wysocki, T., Pierce, J., Caldwell, C., Aroian, K., Miller, L., Farless, R., Hafezzadeh, I., McAninch, T. and Lee, J. (2018). A Web-Based Coping Intervention by and for Parents of Very Young Children With Type 1 Diabetes: User-Centered Design. JMIR Diabetes, 3(4), p.e16. (PubMed)