A Dento-Technical Model for the Analysis and Classification of Palatal Rugae

Abstract

Identification of the persons is a key necessity of forensic science, dentistry, and medico-legal investigation. The traditional methods of identification including fingerprints, dental records, and DNA samples are quite valid but, the methods are not consistently useful in situations involving severe trauma, fire incidents, or severe decomposition. The irregular ridges present on the anterior part of the hard palate are termed as palatal rugae, which are unique, stable and are resistant to postmortem alterations; hence, they make a useful biometric identifier in human identification. It is a dento-technical model of grasping and examining palatal rugae by using a mixture of dental images and computer technologies. The suggested strategy integrates standardized image capture of the dentures, preprocessing of the images, and automated techniques of classifying the images to facilitate better visualization, measurement, and interpretation of rugae patterns. The system reduces the number of manual actions required and biases of the observers and enhances the accuracy and reproducibility by using both traditional machine learning and deep learning classifiers. The system provides increased accuracy and repeatability by combining two classifier methods, namely, traditional machine learning (feature-based morphometric analysis) and deep learning architecture (automated pattern recognition).The paper has shown that the combination of digital image processing and dento-technical analysis is a stable and scalable platform of palatal rugae analysis. The suggested model assists in applying forensic odontology, clinical dentistry, and dental education and thus helps in the development of technology-driven and objective systems of identification.