Healthcare AI, Built from Within: A Human-Centered Approach to Innovation

Clinicians today face increasing cognitive burdens, particularly around chart review, which contribute to burnout and reduced satisfaction. At the same time, patient trust can erode if a provider appears unaware of new diagnoses or medication changes - risks that are heightened for complex patients or those with limited health literacy. This presentation explores how University of Rochester Medical Center built and piloted PreChart Pro, a precharting tool designed to enhance the clinician-patient relationship. PreChart Pro synthesizes structured data and narrative clinical notes into concise, verifiable summaries that help primary care providers prepare for visits more efficiently while maintaining visibility into the source information. The presentation will illustrate a human-centered approach to tool AI development, showing how generative AI can be used to support clinicians and patients - rather than replace or distance them - by prioritizing transparency, clinician oversight, and iterative feedback