MediMovie: Using ChatGPT and Text-to-Video AI to Bring Narrative Medicine into the 21st Century
by Jenae Vancura
Coauthors: Thanks to the MI4 Internship Program, especially Dr. Anthony Chang and Dr. Sharief Taraman, for the introduction to medical innovation in summer 2019, and for their continued guidance and support in the years following.
Medical Devices & Digital Health
Since the launch of its prototype in November 2022, Chat GPT has grown tremendously in popularity, accumulating a 2633% rise in Google searches in the last three months. The software had 18.3 million page views in November 2022 to 304 million in December 2022, and a whopping 1 billion page visits in February 2022 alone1. These statistics point to increasing awareness and acceptance of AI in the day-to-day lives of your average citizens, who are beginning to see how it can be used in a variety of sectors, including healthcare.
Narrative medicine, on the other hand, has often been left behind when it comes to such technological advancements. The two ideas, narrative medicine and medical innovation, are often depicted as juxtaposed if not mutually exclusive entities. There have been many calls for the increased use of narrative medicine in pediatrics, specifically, “to care for [children’s] stories with humility and respect for how they fit in the larger fabric of their families and communities.” 2 In many ways, narrative medicine provides a vehicle through which medical providers can connect with pediatric patients in a way that aligns with their unique understanding of the world. And while the plea to bring narrative medicine into pediatric medicine exists3, there are still questions on how exactly providers can put these models into clinical practice.
This abstract proposes the utilization of Chat GPT and Text-to-Video AI software in the creation of “MediMovie”, a clinical application that will offer AI-generated videos explaining medical diagnoses and treatments in an engaging, age-appropriate manner to pediatric patients.
Users would launch the app and be presented with a variety of prompts for story generation, including gender, age, and diagnosis. Users would also have the option to add first name and/or keywords to customize the story to a child’s interests (e.g. princesses, dinosaurs, superheroes). When the user presses “submit” the movie will begin to generate within a matter of minutes.
MediMovie uses the inputs submitted to generate a ChatGPT Prompt, which uses AI-powered technology to create the fictional story with the user’s preferences. Medi Movie would then utilize Text-to-Video AI technology to convert this story into an animated movie that the patient can watch at any time. (Fictional example can be provided).
Though narrative medicine and medical innovation have long-been thought of as mutually exclusive entities, MediMovie aims to link the two. Through partnering with physicians, parents, and other members of a child’s care team, MediMovie hopes to drive narrative medicine into the 21st Century.