This study aims to determine pelvic anthropometry characteristics and logistic regression formula for adult sex identification obtained from adult three-dimensional pelvic computed tomography images. This study was an observational analytical study with retrospective regression and cross-sectional approach. The population was all patients at Radiology Installation of Dr. Soetomo General Academic Hospital as referral hospital in East Indonesian region, from September to December 2019 who underwent 3D pelvic CT examination. Then, age distribution and pelvic measurements data were obtained. In this case, statistical analysis was conducted for all the data obtained. A number of 204 samples were included in this study. All radiologic components were also significantly different between sexes (p < 0.05) except for transverse diameter of sacral segment (p = 0.180). Moreover, the conjugate pelvic inlet diameter (CPID), the left innominate height (LIH), and sub pubic angle (SPA) showed significant values for regression formula to determine an adult’s sex using 3D pelvic CT. The calculation result > 0 is a prediction for female while < 0 is a prediction for male. From logistic regression model calculation, a high validity value (91.05%) was found with 100% sensitivity to identify male sex and 81.1% specificity to identify female sex. There were differences on radiometric variable characteristics in pelvic anthropometric study among adult Indonesians at Dr. Soetomo General Academic Hospital, Surabaya. The estimated values of pelvic measurements using 3D CT images could develop a pelvic model with a regression formula with high accuracy value using CPID, LIH, and SPA values.

No abstract

Pelvic bones are the most reliable indicator of sex in adults because of its sexual dimorphism. Medical imaging modalities e.g. Computed Tomography (CT) provide data sources to examine modern human variation quantitatively. This study aims to determine pelvic anthropometry characteristics and logistic regression formula for adult sex identification obtained from pelvic 3D CT. This study was an observational analytical study with retrospective regression and cross-sectional approach. The population was all patients in Radiology Installation of Dr. Soetomo General Hospital, Surabaya, Indonesia, from September to December 2019 who underwent pelvic 3D CT examination. Then, age distribution and pelvic measurements data were obtained. In this case, statistical analysis was conducted for all the data obtained. A number of 204 samples were included in this study. Mean age of the patients was 50.23 ± 14.36 years. All radiologic components were also significantly different between sexes (p < 0.05) except for transverse diameter of sacral segment (p = 0.180). Moreover, eta test was performed and found that APOD, CPID, LIH, TPO, and SPA had the strongest correlation. Those variables were used for making statistical models with logistic regression as sex = (0,125 x CPID) – (0,180 x LIH) + (0,078 x SPA) + 8,912. The calculation result > 0 is a prediction for female while < 0 is a prediction for male. From logistic regression model calculation, a high validity value (91.05%) was found with 100% sensitivity to identify male sex and 81.1% specificity to identify female sex. There were differences on radiometric variable characteristics in pelvic anthropometric study. The regression formula for sex determination in adults using 3D-CT pelvic provides a pelvic model sex determination with higher validity and sensitivity for male identification, as well as higher a specificity for female identification

Background Pelvic bones are the most reliable indicator of sex in adults because of its sexual dimorphism. Medical imaging modalities e.g. Computed Tomography (CT) provide data sources to examine modern human variation quantitatively. This study aims to determine pelvic anthropometry characteristics and logistic regression formula for adult sex identification obtained from pelvic 3D CT. Methods This study was an observational analytical study with retrospective regression and cross-sectional approach. The population was all patients in Radiology Installation of Dr. Soetomo General Hospital, Surabaya, Indonesia, from September to December 2019 who underwent pelvic 3D CT examination. Then, age distribution and pelvic measurements data were obtained. In this case, statistical analysis was conducted for all the data obtained. Results A number of 204 samples were included in this study. Mean age of the patients was 50.23 ± 14.36 years. All radiologic components were also significantly different between sexes (p < 0.05) except for transverse diameter of sacral segment (p = 0.180). Moreover, eta test was performed and found that APOD, CPID, LIH, TPO, and SPA had the strongest correlation. Those variables were used for making statistical models with logistic regression as sex = (0,125 x CPID) – (0,180 x LIH) + (0,078 x SPA) + 8,912. The calculation result > 0 is a prediction for female while < 0 is a prediction for male. From logistic regression model calculation, a high validity value (91.05%) was found with 100% sensitivity to identify male sex and 81.1% specificity to identify female sex. Conclusion There were differences on radiometric variable characteristics in pelvic anthropometric study. The regression formula for sex determination in adults using 3D-CT pelvic provides a pelvic model sex determination with higher validity and sensitivity for male identification, as well as higher a specificity for female identification