A skeletal approach to root canal centreline detection from dental micro-CT records

This paper presents a novel image processing procedure dedicated to the automated detection of the medial axis of the root canal from dental micro CT records. The 3D model of root canal is built up from several hundreds of parallel cross sections, using image enhancement and an enhanced fuzzy c-means based partitioning, center point detection in the segmented slice, three dimensional inner surface reconstruction and curve skeleton extraction in three dimensions. The central line of the root canal can finally be approximated as a 3D spline curve. The proposed procedure can help prepare several kinds of endodontic interventions.