EM2607F “Beyond Model Performance: What Kind of Care Does AI Enable?” (IM GR-073126)
This talk will examine how AI is changing the nature of clinical work, patient participation, and healthcare delivery, and why model accuracy alone tells us little about real-world impact. Drawing on experiences from Learning Health Systems, digital health implementation, and human-centered design, I will argue that evaluation must begin long before deployment and continue throughout the innovation lifecycle. Ultimately, the future of AI depends on moving beyond asking whether AI can perform a task and instead ask a more fundamental question: What kind of care does AI enable?
Target Audience
UT Southwestern faculty, fellows, residents and medical students, community
physicians, nurse clinicians, physician assistants and nurses.
Learning Objectives
At the conclusion of this activity, the participant should be able to:
- Define changing roles of stakeholders in healthcare
- Explain why model accuracy is not sufficient in healthcare
- List ways AI could enable improvements in healthcare delivery
Wendy Chapman, Ph.D.
Professor of Internal Medicine
Division of General Internal Medicine
Director, Center for Clinical Informatics
Available Credit
- 1.00 AMA
- 1.00 Attendance
- 1.00 Self-Assessment (MOC)
Price
Required Hardware/software
Activities should be run with recent versions of common browsers, including Internet Explorer, Firefox and Google Chrome

Facebook
X
LinkedIn
Forward