Artificial intelligence helped clinicians to speed up the design of diabetes prevention software, a brand new study finds.
Publishing online March 6 within the Journal of Medical Web Research, the study examined the capabilities of a type of artificial intelligence (AI) called generative AI or GenAI, which predicts likely options for the following word in any sentence based on how billions of individuals used words in context on the web. A side effect of this next-word prediction is that the generative AI “chatbots” like chatGPT can generate replies to questions in realistic language, and produce clear summaries of complex texts.
Led by researchers at NYU Langone Health, the present paper explores the appliance of ChatGPT to the design of a software program that uses text messages to counter diabetes by encouraging patients to eat healthier and get exercise. The team tested whether AI-enabled interchanges between doctors and software engineers could hasten the event of such a personalised automatic messaging system (PAMS).
In the present study, eleven evaluators in fields starting from medicine to computer science successfully used ChatGPT to provide a version of the diabetes tool over 40 hours, where an original, non-AI-enabled effort had required greater than 200 programmer hours.
“We found that ChatGPT improves communications between technical and non-technical team members to hasten the design of computational solutions to medical problems,” says study corresponding creator Danissa Rodriguez, PhD, assistant professor within the Department of Population Health at NYU Langone, and member of its Healthcare Innovation Bridging Research, Informatics and Design (HiBRID) Lab. “The chatbot drove rapid progress throughout the software development life cycle, from capturing original ideas, to deciding which features to incorporate, to generating the pc code. If this proves to be effective at scale it could revolutionize healthcare software design.”
AI as Translator
Generative AI tools are sensitive, say the study authors, and asking an issue of the tool in two subtly alternative ways may yield divergent answers. The skill required to border the questions asked of chatbots in a way that elicits the specified response, called prompt engineering, combines intuition and experimentation. Physicians and nurses, with their understanding of nuanced medical contexts, are well positioned to engineer strategic prompts that improve communications with engineers, and without learning to write down computer code.
These design efforts, nevertheless, where care providers, the would-be users of a brand new software, seek to advise engineers about what it must include will be compromised by attempts to converse using “different” technical languages. In the present study, the clinical members of the team were in a position to type their ideas in plain English, enter them into chatGPT, and ask the tool to convert their input into the type of language required to guide coding work by the team’s software engineers. AI could take software design only thus far before human software developers were needed for final code generation, but the general process was greatly accelerated, say the authors.
“Our study found that chatGPT can democratize the design of healthcare software by enabling doctors and nurses to drive its creation,” says senior study creator Devin Mann, MD, director of the HiBRID Lab, and strategic director of Digital Health Innovation inside NYU Langone Medical Center Information Technology (MCIT).”GenAI-assisted development guarantees to deliver computational tools which are usable, reliable, and in-line with the very best coding standards.”
Together with Rodriguez and Mann, study authors from the Department of Population Health at NYU Langone were Katharine Lawrence, MD, Beatrix Brandfield-Harvey, Lynn Xu, Sumaiya Tasneem, and Defne Levine. Javier Gonzalez,technical lead within the HIBRID Lab, was also a study creator. This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases grant 1R18DK118545-01A1.
