Background and objective <p>The assessment of age, stature, body mass, and sex in minors is important and lacking in forensic fields. This study aimed to analyze the correlation between hand X-ray measurements with age, sex, stature, and body mass, and develop machine learning-based models for biological profiles estimation.</p> Methods <p>This study retrospectively collected 1238 hand X-ray images (610 males and 628 females) aged 3–15 years. A total of 21 features were measured from left-hand X-ray, and nine machine learning models were established for the estimation of age, stature, body mass, and sex, respectively. The SHapley Additive exPlanations (SHAP) method was used for interpreting models.</p> Results <p>The correlations between features and age, stature, body mass are highly significant, with most correlation coefficients exceeding 0.9. Most features show no significant sex difference in the age group of 3-13-year, whereas all features demonstrate sex differences during the ages of 14 and 15.The extra trees and support vector machine models showed superior performance in the estimation of biological profiles. This study achieved a mean absolute error (MAE) of 3.500&#xa0;cm for stature estimation, an MAE of 0.823 years for age estimation, an MAE of 3.278&#xa0;kg for body mass estimation, and an AUC of 0.860 for sex determination. The mean absolute percentage errors of the best models are approximately 10% or lower, indicating that this method is suitable for estimating stature, followed by age and body mass. The SHAP analysis showed that hand length and Metacarpal length emerged as the predominant features across all tasks.</p> Conclusions <p>Our findings provide a valuable reference for biological profiles assessment in minors, particularly in estimating stature, followed by age, body mass, and sex. This method shows potential as a robust and practical tool for the identification of minors in forensic medicine.</p>