Testing the Belle AI Dermatology Image Reference Tool to Support Primary Care Diagnosis of Skin Conditions in Real-World Clinics

What this Trial Is About

Researchers are evaluating a new artificial intelligence AI tool designed to assist primary care providers in assessing skin conditions. This AI-powered dermatology image reference app works with a smartphone to provide reference images, helping providers make diagnoses using their medical training. The study aims to compare diagnoses made by primary care providers using the AI tool to those of a panel of dermatologists, considered the gold standard. The research is supported by the Advanced Research Projects Agency for Health ARPA-H and focuses on improving diagnostic accuracy and care in real-world clinical settings. The study involves primary care providers using the Belle AI tool, which analyzes images captured via a smartphone app and identifies over 2,000 skin conditions based on a large image database. Providers use the app during patient visits to capture images and receive differential diagnoses, which they review before making final decisions. After the initial visit, participants receive notifications to submit follow-up images using the app over a two-week period. Providers reassess diagnoses based on updated AI analysis, and follow-up visits may be scheduled or canceled depending on results. Providers receive training on the app and give feedback through surveys. Participants present with a primary dermatological complaint and consent to study procedures, including using the smartphone app to submit images. Researchers monitor diagnosis accuracy compared to a dermatology review committee consensus and assess economic impact and provider experience. The study lasts about 14 days per participant, with ongoing documentation in electronic medical records and continuous case review to determine care needs or resolution.

Application of the Belle.AI Dermatological Image Reference System for Patient Diagnosis in an Active Clinical Setting