Developing and Validating an AI Colorimetric System for Objective Diagnosis of Peri-Implant Mucosal Redness and Its Effect on Clinician Accuracy

What this Trial Is About

Researchers are evaluating an artificial intelligence AI-based colorimetric system to improve the diagnosis of peri-implant mucosal erythema, a key sign of inflammation around dental implants. This study aims to develop and validate objective tools including a color index and AI models that automatically detect and localize redness from digital intraoral scan images. The study also assesses how AI assistance affects the accuracy and confidence of clinicians with different levels of experience. The study involves adult patients who have received intraoral scans of their dental implants using a specialized scanner. The AI system is developed and tested in three phases initial development using internal scan data, external validation on a separate patient group, and a clinical utility phase where clinicians diagnose images both with and without AI help. The AI provides a color index value and highlights suspected inflamed areas on images to support diagnosis. Participants include patients contributing their de-identified dental images for AI training and validation, and clinicians who complete diagnostic tests using these images. Researchers measure the accuracy of AI detection, agreement with expert diagnoses, and improvements in clinician performance when aided by AI. The main outcome is the diagnostic accuracy of the AI system measured about three months after study start. The study duration and participant involvement vary across phases, with ongoing assessments of diagnostic reliability and utility.

AI-Assisted Colorimetric Diagnosis of Peri-Implant Mucosal Erythema