Back

Statistical analysis of sexual dimorphism in adults of Western UP by Computed Tomography based Sternal Morphometry

Dixit, S.; Singh, H. K.; Kumar, R.

2026-08-03 radiology and imaging
10.64898/2026.07.28.26359105 medRxiv
Show abstract

Background: Sternal dimensions show population-specific sexual dimorphism and may support forensic identification when more informative skeletal elements are unavailable. Computed tomography (CT) permits non-destructive, reproducible three-dimensional assessment of the sternum. Objective: To quantify sex- and age-associated differences in CT-derived sternal measurements among adults from Western Uttar Pradesh, India. Methods: This prospective cross-sectional study included 250 adults aged 20-80 years (155 males and 95 females) referred for chest CT at a tertiary hospital in Moradabad. Images were acquired using a Philips Ingenuity Core 128-slice scanner with 1-mm sections and 0.5-mm reconstruction increments. Manubrium width (MW), manubrium length (ML), sternal body length (B), corpus sterni widths at the first and third sternebrae (CSWS1 and CSWS3), sternal index (SI), combined length (CL), and sternal area (SA) were evaluated. Independent-samples t tests, one-way analysis of variance with Tukey post hoc testing, and Pearson correlations were used. Results: Mean age was 48.1 years. Males had greater MW (57.3 vs 49.9 mm), ML (45.3 vs 40.4 mm), B (90.2 vs 75.7 mm), CSWS1 (26.1 vs 23.0 mm), CSWS3 (31.6 vs 27.9 mm), CL (135.5 vs 116.1 mm), and SA (5258.3 vs 3880.5 mm; all p < 0.001). SI was greater in females (53.8 vs 50.8; p = 0.009). Across age groups, only MW (p = 0.001) and SA (p = 0.002) differed significantly. Strong correlations were observed for B with CL (r = 0.902), CL with SA (r = 0.895), B with SA (r = 0.869), and MW with SA (r = 0.847; all p < 0.001). Conclusion: CT-derived sternal morphometry demonstrates marked sexual dimorphism in this Western Uttar Pradesh sample, particularly for SA, CL, B, and MW. These reference values provide a basis for population-specific forensic models; however, predictive accuracy cannot be inferred without classifier development and validation.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.