Digital intraoral scans of upper teeth used to estimate sex and stature in 320 adults from North Maharashtra

Researchers tested whether three-dimensional measurements taken from intraoral scans of teeth can help estimate sex and stature in adults from North Maharashtra, India. The cross-sectional study, published in the Egyptian Journal of Forensic Sciences, included 320 healthy adults (160 men and 160 women, mean age 23.2 ± 3.03 years) selected by stratified random sampling biomedcentral.comEgyptian Journal of Forensic Sciencesbiomedcentral.comOpen source ↗. According to the authors, generalized standards may introduce systematic bias in heterogeneous populations such as India, which creates a need for region-specific models.

Measurements and sample

Participants had to have fully erupted, caries-free, unrestored permanent canines and first molars. Stature was measured with a standardized stadiometer. Digital impressions were obtained with an intraoral scanner (a TRIOS 3, as described in the publication's coverage), and the scans were measured in three-dimensional software.

The team measured maxillary and mandibular intercanine width, canine mesiodistal and buccolingual widths, and the cumulative mesiodistal width of six anterior teeth. Sex differences were assessed with t-tests or Mann–Whitney U tests. Stature was analyzed with Spearman's correlation and multiple linear regression, and sex classification with binary logistic regression and ROC analysis.

Results for sex and stature

Significant sexual dimorphism appeared only in the maxillary parameters. Males had larger intercanine width (p = 0.003), canine mesiodistal width (p = 0.001), canine buccolingual width (p = 0.049) and cumulative anterior mesiodistal width (p = 0.010). Mandibular parameters did not differ significantly biomedcentral.comEgyptian Journal of Forensic Sciencesbiomedcentral.comOpen source ↗.

Stature correlated positively with maxillary intercanine width (r = 0.26, p = 0.001), maxillary canine mesiodistal width (r = 0.22, p = 0.005) and cumulative anterior width (r = 0.13, p = 0.012). Linear regression indicated a 0.79 cm increase in stature per 1 mm increase in maxillary intercanine width (p = 0.002). The final model explained 44% of stature variance in the testing dataset.

Logistic regression showed moderate discriminatory performance for sex classification, with AUC = 0.841 on the training data and 0.782 on the testing data. The authors conclude that the population-specific models may serve as supportive tools for forensic identification when dental remains are available, and that three-dimensional digital measurement improves reproducibility.