Artificial intelligence versus semi-automatic segmentation of the inferior alveolar canal on cone-beam computed tomography scans: A pilot study

Issa, Julien [UCL] Kulczyk, Tomasz Rychlik, Michal Czajka-Jakubowska Agata Olszewski, Raphaël [UCL] Dyszkiewicz-Konwinska, Marta

Background: The inferior alveolar canal (IAC) is a fundamental mandibular structure. It is important to conduct a precise pre-surgical evaluation of the IAC to prevent complications. Recently, the use of artificial intelligence (AI) has demonstrated potential as a valuable tool for dentists, particularly in the field of oral and maxillofacial radiology. Objectives: The aim of the study was to compare the segmentation time and accuracy of AI-based IAC segmentation with semi-automatic segmentation performed by a specialist. Material and methods: Thirty individual IACs from 15 anonymized cone-beam computed tomography (CBCT) scans of patients with at least 1 lower third molar were collected from the database of Poznan University of Medical Sciences, Poland. The IACs were segmented by a trainee in the field of oral and maxillofacial radiology using a semi-automatic method and automatically by an AI-based platform (Diagnocat). The resulting segmentations were overlapped with the use of Geomagic Studio, reverse engineering software, and then subjected to a statistical analysis. Results: The AI-based segmentation closely matched the semi-automatic method, with an average deviation of 0.275 ±0.475 mm between the overlapped segmentations. The mean segmentation time for the AI-based method (175.00 s) was similar to that of the semi-automatic method (175.67 s). Conclusions: The results of the study indicate that AI-based tools may offer a reliable approach for the segmentation of the IAC in the context of dental pre-surgical planning. However, further comprehensive studies are required to compare the methods and consider their limitations more comprehensively.