Prediction of patient disposition: comparison of computer and human approaches and a proposed synthesis
This on-demand webinar does not offer CE credit.
Lead author Yuval Barak-Corren, MD, MS, discusses this month’s JAMIA Journal Club selection:
Barak-Corren Y, Agarwal I, Michelson KA, et al. Prediction of patient disposition: comparison of computer and human approaches and a proposed synthesis [published online ahead of print, 2021 May 9]. J Am Med Inform Assoc. 2021;ocab076. doi:10.1093/jamia/ocab076
Watch the Recording
Yuval Barak-Corren, MD, MS, is a researcher at the Predictive Medicine Group at Boston Children’s Hospital. His work focuses on the intersection of computer science and healthcare and includes the development of advanced prediction models for both clinical and operational purposes. In addition to his research work, Yuval is a practicing pediatrician and is currently pursuing a fellowship in pediatric cardiology at the Schneider Children's Medical Center of Israel. Yuval also serves as a consultant for startups and for the Israeli ministry of health on topics of digital health.
Statement of Purpose
Computer models have been developed to predict patient disposition from the emergency department (ED), yet it is not clear if their ability outperforms that of experienced ED clinicians. We sought to compare the accuracy of computer vs. human physician predictions, to study the factors driving each prediction, and to explore the potential synergies of hybrid physician-computer models.
The target audience for this activity is professionals and students interested in health informatics.
The general learning objective for all of the JAMIA Journal Club webinars is that participants will
- Use a critical appraisal process to assess article validity and to gauge article findings' relevance to practice
After participating in this webinar, the listener should be able to:
- Understand the strengths and weaknesses of both machine-learning and clinician derived predictions, in the context of predicting hospitalizations from the ED.
- Learn ways to integrate computer and clinician predictions, and to acquire better understanding of the added value of such hybrid models.