Developing a Novel Approach for Periapical Dental Radiographs Segmentation

TL;DR

This paper introduces an automated, two-stage method for segmenting teeth in Periapical dental X-ray images, aiding in dental diagnostics and bone lesion detection with high accuracy.

Contribution

The paper presents a novel two-stage algorithm combining pre-processing and integral projection for effective teeth segmentation in dental X-rays.

Findings

High accuracy in teeth segmentation demonstrated

Robustness of the algorithm across different images

Potential application in bone lesion detection

Abstract

Image processing techniques has been widely used in dental researches such as human identification and forensic dentistry, teeth numbering, dental carries detection and periodontal disease analysis. One of the most challenging parts in dental imaging is teeth segmentation and how to separate them from each other. In this paper, an automated method for teeth segmentation of Periapical dental x-ray images which contain at least one root-canalled tooth is proposed. The result of this approach can be used as an initial step in bone lesion detection. The proposed algorithm is made of two stages. The first stage is pre-processing. The second and main part of this algorithm calculated rotation degree and uses the integral projection method for tooth isolation. Experimental results show that this algorithm is robust and achieves high accuracy.