NCT07505485

Artificial Intelligence-Based Assessment of Endosseous Lesions

Artificial Intelligence-Based Assessment of Endosseous Lesions: A Prospective Clinical Study

• Status: Recruiting
• Phase: N/A
• Study type: Interventional
• Enrollment: 10 (estimated)
• Sponsor: University of Bari Aldo Moro · Academic / Other
• Sex: All
• Age: 18 Years – 80 Years
• Healthy volunteers: Accepted

Summary

Despite these advances, CBCT interpretation remains largely qualitative and dependent on the clinician's experience. Conventional evaluation is based on two-dimensional slices and linear measurements, which may underestimate lesion complexity and spatial distribution. Recent developments in Artificial Intelligence in Medicine have introduced automated image segmentation tools capable of identifying lesion boundaries and calculating volumetric data. These technologies allow a transition from subjective assessment to objective, reproducible quantification. The potential clinical advantages include: * Objective measurement of lesion size (volume in mm³) * Improved surgical planning * Enhanced prediction of anatomical involvement * Reduction of diagnostic errors * Standardization of follow-up and outcome assessment Therefore, the aim of the present study was to evaluate the clinical impact of AI-based segmentation and volumetric analysis of endosseous lesions compared to conventional CBCT interpretation.

Conditions

• Maxillary Cyst
• Mandibular Cyst

Interventions

Type	Name	Description
DIAGNOSTIC_TEST	AI assisted Evaluation	CBCT scans were processed using AI-based software capable of: * Automated segmentation of the lesion * 3D reconstruction * Volumetric calculation

Timeline

• Start date: 2026-04-01
• Primary completion: 2026-05-01
• Completion: 2026-05-01
• First posted: 2026-04-01
• Last updated: 2026-04-15

Locations

2 sites across 1 country: Italy