Stature estimation from skeletal remains is a cornerstone of forensic identification, complementing other biological attributes such as sex and age. Because stature estimation methods are population-specific, the lack of contemporary reference data for the Turkish population represents a significant limitation in forensic anthropology.

This retrospective study included 102 adults (51 males, 51 females) aged 20–87 years with known stature. Three-dimensional reconstructions derived from computed tomography (CT) scans were used to obtain measurements of the femur, tibia, and fibula. Linear regression models were developed for pooled and sex-specific samples. Model performance was evaluated using 1,000 bootstrap resamples and leave-one-out cross-validation (LOOC).

Among univariate models, fibula length demonstrated the highest predictive accuracy in the pooled sample ( R ² = 0.526; SEE = 5.5 cm). In males, femoral length yielded the lowest estimation error (SEE = 4.76 cm), whereas in females, fibula length provided the best performance (SEE = 5.3 cm). Multivariate models did not result in meaningful improvements in estimation accuracy.

CT-based regression equations derived from lower limb long bones provide reliable stature estimates for the contemporary Turkish population. Expanding sample size and geographic representation may further strengthen population-specific forensic standards in Turkey.