Zulal Oner, Muhammed Kamil Turan2, Serkan Oner, Yusuf Secgin1

Gender prediction with parameters obtained from pelvis computed tomography images and decision tree algorithm

Gender prediction is among the most critical topics in forensic medicine and anthropology since it is the basis of identity (height, weight, ancestry, age). Today, osteometry which is a low-cost, easily accessible method that requires no expertise is preferred when compared to DNA technology, which has several disadvantages such as high cost, accessibility, laboratory facilities, and expert personnel requirements. The Computed Tomography (CT) method, which is little affected by orientation and provides reconstruction opportunities, was selected instead of traditional methods for osteometry. This study aims to predict high and accurate gender with the Decision Tree (DT) algorithms used in the field of health recently. In the present study, CT images of 300 individuals (150 females, 150 males) without a pathology on the pelvic skeleton and between the ages of 25 and 50 were transformed into orthogonal form, landmarks were placed on promontorium, sacroiliac joint, iliac crest, terminal line, anterior superior iliac spine, anterior inferior iliac spine, greater trochanter, obturator foramen, lesser trochanter, femoral head, femoral neck, the body of femur, ischial tuberosity, acetabulum, and pubic symphysis, and the coordinates of these landmarks were determined. Then, parameters such as angle and length were obtained with various combinations. These parameters were analyzed with the DT algorithm.The analysis conducted with the DT algorithm revealed that accuracy (Acc) was 0.93, sensitivity was 0.95, specificity was 0.90, and the Matthews correlation coefficient was 0.86 for the pelvic skeleton. It was observed that the accuracy was quite high and more realistic when determined with the DT algorithm. In conclusion, the DT algorithm with multiple parameters and samples on pelvic CT images could improve the Acc of gender prediction.

PDF

___

Colman KL, van der Merwe AE, Stull KE, et al. The accuracy of 3D virtual bone models of the pelvis for morphological sex estimation. Int J Legal Med. 2019;133:1853-60.
Giurazza F, Schena E, Del Vescovo R, et al. Sex determination from scapular length measurements by CT scans images in a Caucasian population. Conf Proc IEEE Eng Med Biol Soc. 2013;2013:1632-5.
Sinhorini PA, Costa IAP, Lopez-Capp TT, et al. Comparative analysis of four morphometric methods for sex estimation: A study conducted on human skulls. Leg Med (Tokyo). 2019;39:29-34.
Ahmed AA. Estimation of sex from the upper limb measurements of Sudanese adults. J Forensic Leg Med. 2013;20:1041-7.
Bonczarowska JH, Bonicelli A, Papadomanolakis A, et al. The posterior portion of the ilium as a sex indicator: A validation study. Forensic Sci Int. 2019;294:1- 6.
Kim DI, Kwak DS, Han SH. Sex determination using discriminant analysis of the medial and lateral condyles of the femur in Koreans. Forensic Sci Int. 2013;233:121-5.
Best KC, Garvin HM, Cabo LL. An investigation into the relationship between human cranial and pelvic sexual dimorphism. J Forensic Sci. 2018;63:990-1000.
Blake KAS, Hartnett-McCann K. Metric assessment of the pubic bone using known and novel data points for sex estimation. J Forensic Sci. 2018;63:1472-8.
Karakas HM, Harma A, Alicioglu B. The subpubic angle in sex determination: anthropometric measurements and analyses on Anatolian Caucasians using multidetector computed tomography datasets. J Forensic Leg Med. 2013;20:1004-9.
Hayashizaki Y, Usui A, Hosokai Y, et al. Sex determination of the pelvis using Fourier analysis of postmortem CT images. Forensic Sci Int. 2015;246:1-9.
Climent MT, Pardo J, Munoz-Almaraz FJ, et al. Decision tree for early detection of cognitive impairment by community pharmacists. Front Pharmacol. 2018;9:1232.
. Song YY, Lu Y. Decision tree methods: applications for classification and prediction. Shanghai Arch Psychiatry. 2015;27:130-5.
Wankhede KP, Bardale RV, Chaudhari GR, et al. Determination of sex by discriminant function analysis of mandibles from a Central Indian population. J Forensic Dent Sci. 2015;7:37-43.
Oner Z, Turan MK, Oner S, et al. Sex estimation using sternum part lenghts by means of artificial neural networks. Forensic Sci Int. 2019;301:6-11.
d'Oliveira Coelho J, Curate F. CADOES: An interactive machinelearning approach for sex estimation with the pelvis. Forensic Sci Int. 2019;302:109873.
Savall F, Faruch-Bilfeld M, Dedouit F, et al. Metric sex determination of the human coxal bone on a virtual sample using decision trees. J Forensic Sci. 2015;60:1395-400.
Pretorius E, Steyn M, Scholtz Y. Investigation into the usability of geometric morphometric analysis in assessment of sexual dimorphism. Am J Phys Anthropol. 2006;129:64-70.
. McBride DG, Dietz MJ, Vennemeyer MT, et al. Bootstrap methods for sex determination from the os coxae using the ID3 algorithm. J Forensic Sci. 2001;46:427-31.
Steyn M, Iscan MY. Metric sex determination from the pelvis in modern Greeks. Forensic Sci Int. 2008;179:1-6.
Turan MK, Oner Z, Secgin Y, et al. A trial on artificial neural networks in predicting sex through bone length measurements on the first and fifth phalanges and metatarsals. Comput Biol Med. 2019;115:103490.