Johns Hopkins has big plans for AI in Epic chart summarization

Dr. Brian Hasselfeld of Johns Hopkins Medicine delves into the institution's use of AI technologies, including ambient scribing, patient portal message drafting, and chart summarization in collaboration with Epic.

Ambient Scribe Technology: Hasselfeld discusses the implementation of ambient AI technology, which captures clinical encounters and generates draft notes using large language models like GPT. This technology aims to alleviate documentation burdens on clinicians and improve patient interactions by allowing practitioners to focus more on patients.

Patient Portal Message Draft Replies: Johns Hopkins Medicine employs AI for drafting responses to patient portal messages, addressing the surge in patient communication during the pandemic. Clinicians can select draft responses generated by the AI model, reducing cognitive burden and in-basket burnout, though applicability remains variable.

Chart Summarization via AI: Johns Hopkins is in early stages of developing AI for chart summarization, initially focusing on inpatient hospital course summaries. The aim is to extract relevant information from unstructured data to provide concise summaries, particularly during inpatient admissions, to streamline handoffs and improve efficiency.

Overall, these AI initiatives aim to enhance clinician workflows, improve patient interactions, and address challenges in healthcare documentation and information management. Collaboration with Epic underscores the commitment to leveraging technology for improving healthcare delivery.