Skip to content
DataLookout
Menu

NCT07907900

Diagnostic Accuracy of Large Language Models for Peri-Implant Marginal Bone Loss on Panoramic Radiographs

StatusNot Yet Recruiting
PhaseNot specified
Primary completion (estimated)
Track this trial

Free. We email you which fields changed when ClinicalTrials.gov updates this record.

Record History

All 1 version: 1 with changes below.

  1. Study First Submitted

Endpoints

Primary

Sensitivity and specificity of each large language model for detecting peri-implant marginal bone loss
Timeframe: Single time point, at completion of model evaluation (approximately 8 weeks after study start); retrospective images, no follow-up

Secondary (4)

Positive predictive value, negative predictive value and overall accuracy of each model
Agreement between model severity grading and the reference standard
Area under the ROC curve of each model
Inter-rater reliability of the three blinded clinicians

Arms and interventions

Implant sites with marginal bone loss

Anonymized, standardized region-of-interest (ROI) images of functionally loaded dental implants on panoramic radiographs showing peri-implant marginal bone loss (Froum and Rosen Score 1, 2 or 3) according to the consensus of three blinded clinicians. Target: 392 implant sites.

Implant sites without marginal bone loss

Anonymized, standardized ROI images of functionally loaded dental implants on panoramic radiographs with no visible peri-implant marginal bone loss (Froum and Rosen Score 0) according to the consensus of three blinded clinicians. Target: 392 implant sites.

Diagnostic tests: Multimodal large language model evaluation (ChatGPT, Gemini, Claude, DeepSeek, Grok)

Sites

1 site in 1 country

  • Ankara University Faculty of DentistryAnkara, Turkey (Türkiye)