Using Belle.ai Skin Image Tool to Help Doctors Diagnose Common Skin Diseases Study With Patients at Advocate Health Dermatology Clinics

What this Trial Is About

This research aims to evaluate the accuracy of the Belle.ai software in diagnosing common dermatologic diseases across different skin tones. The software uses deep learning technology to analyze clinical images of skin conditions and provides a differential diagnosis from a database of over 2,000 skin conditions based on more than 500,000 images. The study will determine how well the AI-generated diagnoses match those of dermatology experts. Participants who consent will have three images taken of their skin condition using the Belle.ai software. These images will be de-identified and uploaded into the system, where the software will generate a single list of possible diagnoses. Dermatology experts will then review the cases to compare their diagnoses with the AI results. The study will be conducted at dermatology clinics within the Advocate Health system. During the study, images will be captured and uploaded securely without any personal identifiers. The Dermatologic Review Committee will assess the concordance between the AIs primary diagnosis and expert opinions, aiming for greater than 80% agreement. Participants will not need to complete any additional forms, and their involvement primarily involves image capture during their clinic visit. The main outcome measured is the agreement between AI and physician diagnoses on the day of image capture.

A Skin Image Reference Tool to Aid Healthcare Providers' Diagnosis