Predictive models using data mining techniques such as machine learning (ML) algorithms are being developed for use in clinical medicine with increasing frequency. Despite prolific publication in the scientific literature, there is little known about how academic medical centers (AMCs) develop, implement, and maintain these models from an operational perspective. Using qualitative analysis of structured interviews, we present here the current state of ML models in clinical care at peer AMCs across the US in terms of challenges encountered and recommended best practices.
Describe the new knowledge and additional skills the participant will gain after attending your presentation.: Practical knowledge of the numerous challenges academic medical centers face in adopting machine learning algorithms for clinical care as well as specific best practices.
Anisha Chandiramani (Presenter)
Joshua Watson, Duke University
Carolyn Hutyra, Duke University
Shayna Clancy, Duke University
Kumar Ilangovan, Duke University
Nancy Nderitu, Duke University
Armando Bedoya, Duke University
Eric Poon, Duke University