CMU logo
Search
Expand Menu
Close Menu

HCI Behind the Hospital Doors: EHR Challenges and Human Machine-Learning Interaction

Open in new window

Speaker
Susan Regli
Human Factors Scientist, University of Pennsylvania Health System

When
-

Where
Newell-Simon Hall 1305 (Michael Mauldin Auditorium)

Video
Video link

Description

This talk will address two aspects of hospital systems that sorely need attention from the human-computer interaction community. The first aspect, electronic health record usability and safety, is often treated as a vendor problem but poses unique and critical challenges when the EHR is implemented and evolving in a hospital system. I will provide examples from an EHR implementation within an academic medical system that capitalized on HCI methodologies, as well as efforts since system go-live to enable ongoing excellence in EHR informatics.

The second aspect, predictive healthcare, is an area of innovation that has arisen out of the availability of enormous amounts of discrete data from EHRs. Moving from machine learning predictions of patient risk to actual programs to improve outcomes requires a strong integration between data science and clinical expertise, with HCI as the glue. I will discuss elements needed to advance a nascent field of “human machine-learning interaction” and describe the data science “mad libs” process we have developed to create successful applied predictive healthcare programs in partnership with innovative clinicians.

Speaker's Bio

Dr. Susan Harkness Regli is the human factors scientist in the Clinical Effectiveness and Quality Improvement department of the University of Pennsylvania Health System. As part of the Penn Value Improvement effort, she works closely with clinical informatics specialists and the Chief Medical Information Officer to promote best practices in human-computer interaction design and usability for health information technology. She also provides user-centered design leadership to connect clinicians and data scientists in developing effective predictive healthcare projects using machine learning.

Dr. Regli previously worked at Lockheed Martin, Xerox PARC, and VerticalNet, Inc. Her research has included user-centered design for multimodal and natural-language understanding technology; electronic mail in computer supported collaborative work; mobile computing applications for intelligence collection and exchange; and enhancements to medical information collection and usage. She has published on these professional topics as well as an article on technical writing for the textbook Professional Writing and Rhetoric: Readings from the Field and a book of poetry called Her Arms Can Hold Me. Dr. Regli earned her doctorate in Rhetoric from Carnegie Mellon University, where she was affiliated with the Human-Computer Interaction Institute. She also holds degrees from St. Joseph’s University, the University of Virginia, and Hollins College.

Host
Chris Neuwirth