Description
Electronic medical record (EMR) systems typically present patient data with little prioritization leading to the risk of information overload. We developed a learning EMR system that uses machine learning models to predict which data are relevant in a patient and highlight them to the physician. Building models require large amounts of which patient data are accessed by physicians. We developed an eye-tracking system as a potentially viable method to automatically capture such data.
Describe the new knowledge and additional skills the participant will gain after attending your presentation.: After attending the presentation, the participant will gain knowledge about the problem of information overload in EMRs and a solution to this problem called the learning EMR system that uses machine learning to identify relevant EMR data to highlight for a given patient and clinical context. It does so by analyzing patterns of patient data access in past patients in the EMR and building machine learning models to predict which data to highlight. Building models require large amounts of which patient data are accessed by physicians. An inexpensive eye-tracking system is a potentially viable method to automatically capture such data.
Authors:
Shyam Visweswaran (Presenter)
University of Pittsburgh
Presentation Materials: