Please use this identifier to cite or link to this item:
Title: Use of deep learning in forensic sex estimation of virtual pelvic models from the Han population
Authors: Cao, Yongjie
Ma, Yonggang
Yang, Xiaotong
Xiong, Jian
Wang, Yahui
Zhang, Jianhua
Qin, Zhiqiang
Chen, Yijiu
Vieira, Duarte Nuno 
Chen, Feng
Zhang, Ji
Huang, Ping
Keywords: Forensic sciences; forensic anthropology; sex estimation; pelvis; deep learning; convolutional neural network
Issue Date: 2022
Project: National Natural Science Foundation of China (81801873, 81722027, 81671869 82072115 and 81922041) 
grants from the Ministry of Finance (No. GY2020G-2) 
Science and Technology Commission of Shanghai Municipality (No. 17DZ2273200 and 19DZ2292700). 
Serial title, monograph or event: Forensic Sciences Research
Abstract: Accurate sex estimation is crucial to determine the identity of human skeletal remains effectively. Here, we developed convolutional neural network (CNN) models for sex estimation on virtual hemi-pelvic regions, including the ventral pubis (VP), dorsal pubis (DP), greater sciatic notch (GSN), pelvic inlet (PI), ischium, and acetabulum from the Han population and compared these models with two experienced forensic anthropologists using morphological methods. A Computed Tomography (CT ) dataset of 862 individuals was divided into the subgroups of training, validation, and testing, respectively. The CT -based virtual hemi-pelvises from the training and validation groups were used to calibrate sex estimation models; and then a testing dataset was used to evaluate the performance of the trained models and two human experts on the sex estimation of specific pelvic regions in terms of overall accuracy, sensitivity, specificity, F1 score, and receiver operating characteristic (ROC) curve. Except for the ischium and acetabulum, the CNN models trained with the VP, DP, GSN, and PI images achieved excellent results with all the prediction metrics over 0.9. All accuracies were superior to those of the two forensic anthropologists in the independent testing. Notably, the heatmap results confirmed that the trained CNN models were focused on traditional sexual anatomic traits for sex classification. This study demonstrates the potential of AI techniques based on the radiological dataset in sex estimation of virtual pelvic models. The excellent sex estimation performance obtained by the CNN models indicates that this method is valuable to proceed with in prospective forensic trials.
ISSN: 2096-1790
DOI: 10.1080/20961790.2021.2024369
Rights: openAccess
Appears in Collections:FMUC Medicina - Artigos em Revistas Internacionais

Show full item record

Google ScholarTM




This item is licensed under a Creative Commons License Creative Commons