Evaluation of age by Kvaal’s modified measurements (KMM) using computer-aided imaging software and digitized parameters

In the field of forensic science, identification of the unknown is the primary task to be done by forensic experts. Age assessment is one of the most important biological parameters of identification. The durability of teeth makes them one of the most suitable evidence for age estimation.The present study was performed to develop and validate new teeth specific multiple regression models for age estimation based on Kvaal’s modified measurements (KMM) and to check whether this approach leads to statistically comparable and reproducible results or not.Sample comprised of 1280 RVGs of 4 teeth per subject (maxillary canine, maxillary second premolar, mandibular canine, and mandibular first premolar) collected from 320 individuals (160 males and 160 females) having the age range 15–54 years. Thirteen radiographic measurements (3 vertical and 10 horizontal) were performed on all radiographs using computer-aided software (Adobe Photoshop). Various ratios from these vertical as well as horizontal measurements were calculated. These ratios were subjected to multiple regressions. Developed multiple regression models were validated for their accuracy on a different set of 80 RVGs obtained from another 20 subjects of known age. Comparative analysis was done to check the difference between the mean chronological age and estimated age.It was found that the pulpal width showed a better correlation with the chronological age when contrasted with vertical ratios. The concordance correlation coefficient was used to check the intraobserver reproducibility of measurements. The multiple regression models were established. The determination coefficients (r²) of the models for all the studied tooth varies from 0.331 to 0.631 with SEE varies from 6.27 to 7.89 years. Results of the present study suggested a significant negative correlation between chronological age and variables. Maxillary second premolar demonstrated the most noteworthy value of r².The results of the present study give an inference for the feasibility of this technique. Our study indicates that the width measurements are age-dependent variable in adults as these ratios showed a better correlation with age. High resolution of RVGs has improved the age estimation error rate.