Social Media & Technology
EHR/Medical Informatics
Trainee
Srinivasan Suresh, MD, MBA
Vice President & CMIO; Professor of Pediatrics
UPMC Childrens Hospital of Pittsburgh
Pittsburgh, Pennsylvania, United States
Sriram Ramgopal, MD (he/him/his)
Assistant Professor
Pediatrics
Ann & Robert H. Lurie Children's Hospital of Chicago
Chicago, Illinois, United States
Judith Dexheimer, PhD, MBA (she/her/hers)
Associate Professor
Emergency Medicine, Biomedical Informatics
Cincinnati Children's Hospital Medical Center
Cincinnati, Ohio, United States
Armand Antommaria, MD, PhD (he/him/his)
Director, Ethics Center
Cincinnati Children's Hospital Medical Center
Cincinnati, Ohio, United States
Workshop
Description: Generative artificial intelligence (AI) refers to AI models and systems that have the capability to generate new, creative content, by learning from existing data. Large language models (LLMs), such as ChatGPT can create immense value in pediatrics, including in clinical care, research, education, patient engagement, and practice management. LLMs may enhance care by providing immediate access to comprehensive medical
knowledge and easily understandable decision aids. For research, LLMs can help in the identification of research topics, data analysis, and literature reviews. LLMs can enhance educational approaches via the generation of objective structured clinical examinations (OSCE). LLMs can improve communication with caregivers and patients by offering clear and accessible explanations of medical procedures and conditions,
thereby reducing anxiety and ensuring better understanding. Embedded into an electronic health record system and existing care workflows, LLMs can assist in clinical documentation and in triaging and discharging patients.
Attendees will interact with our experienced workshop leaders (clinicians, clinical researcher, informatician, and ethicist) and do a deep dive of LLMs in the care of children, specifically focusing on increasing efficiency and productivity of pediatricians while simultaneously reducing the administrative burden. Attendees will develop familiarity of the value of LLMs in research and clinical care, including developing appropriate query prompts. In addition to ethical, regulatory, privacy and security issues, there is substantial concern that LLMs perpetuate bias and contribute to disparities. The workshop leaders and attendees will collectively explore strategies to mitigate the potential risks of LLMs in pediatrics.