Abstract

Sex determination from skeletal remains is crucial in forensic anthropology, aiding in medicolegal investigations and archaeological studies. Conventional methods rely on bone morphology, but alternative, non-invasive approaches are needed. This study explores the utility of maxillary sinus dimensions, measured from plain CT scans, for sex determination in a Northern Indian population. A cross-sectional study was conducted with 400 participants, evenly distributed by sex. Maxillary sinus dimensions were measured using specialized software. Statistical analysis included t-tests, ROC analysis, cross-validation, and assessment of age-related changes and reliability. Maxillary sinus dimensions exhibited sexual dimorphism, with males having larger dimensions. ROC analysis yielded high AUC values (0.85 for MSH, 0.78 for MSW, and 0.81 for MSL), indicating excellent discrimination between sexes. Cross-validation demonstrated reliable classification, with Random Forest achieving 89.7% accuracy. Age had a significant influence on dimensions. Intra- and inter-rater reliability was excellent. Maxillary sinus dimensions from plain CT scans show promise as a reliable and non-invasive method for sex determination in Northern Indian populations. While age should be considered, these dimensions can enhance forensic anthropology and medicolegal investigations, improving accuracy and efficiency.

Description