Accurate segmentation of digitized dental X-ray records

Abstract

Recent disasters have emphasized the significance of automated dental identification systems. Statistics show that 20% of the 9/11 victims, identified in the first year, were manually identified using dental records. Moreover, 75% of Tsunami victims in Thailand were similarly identified using dental records, compared to 0.5% identified using DNA. This paper addresses the first important problem of an Automated Dental Identification System (ADIS). This system matches image features extracted from multiple dental radiographic records. Dental radiograph record of an individual usually consists of radiographic films. It is an essential step to accurately segment these films from their constituent dental records in order to extract the dental features and achieve high level of automated postmortem identification. In this paper, we propose an automated approach to the problem of segmenting films from their dental records. Challenges include the variability in the background of the dental records including its gray intensity and texture, and variation in the number of films and their dimensions. Our three-stage approach is based on concepts of thresholding, connectivity, and mathematical morphology. We show by experimental evidence that our approach achieves 92% accuracy compared to 74% using previous work suggested in the literature.