Sex Estimation of Medial Aspect of the Ischiopubic Ramus in Adults Based on Deep Learning

doi:10.12116/j.issn.1004-5619.2022.220505

Abstract

Abstract:

Objective To investigate the reliability and accuracy of deep learning technology in automatic sex estimation using the 3D reconstructed images of the computed tomography （CT） from the Chinese Han population. Methods The pelvic CT images of 700 individuals （350 males and 350 females） of the Chinese Han population aged 20 to 85 years were collected and reconstructed into 3D virtual skeletal models. The feature region images of the medial aspect of the ischiopubic ramus （MIPR） were intercepted. The Inception v4 was adopted as the image recognition model, and two methods of initial learning and transfer learning were used for training. Eighty percent of the individuals’ images were randomly selected as the training and validation dataset, and the remaining were used as the test dataset. The left and right sides of the MIPR images were trained separately and combinedly. Subsequently, the models’ performance was evaluated by overall accuracy, female accuracy, male accuracy, etc. Results When both sides of the MIPR images were trained separately with initial learning, the overall accuracy of the right model was 95.7%, the female accuracy and male accuracy were both 95.7%; the overall accuracy of the left model was 92.1%, the female accuracy was 88.6% and the male accuracy was 95.7%. When the left and right MIPR images were combined to train with initial learning, the overall accuracy of the model was 94.6%, the female accuracy was 92.1% and the male accuracy was 97.1%. When the left and right MIPR images were combined to train with transfer learning, the model achieved an overall accuracy of 95.7%, and the female and male accuracies were both 95.7%. Conclusion The use of deep learning model of Inception v4 and transfer learning algorithm to construct a sex estimation model for pelvic MIPR images of Chinese Han population has high accuracy and well generalizability in human remains, which can effectively estimate the sex in adults.

Key words: forensic anthropology, deep learning, sex estimation, three-dimensional reconstruction, pelvis, medial aspect of the ischiopubic ramus, transfer learning

CLC Number:

DF795.6

Yong-gang MA, Yong-jie CAO, Yi-hua ZHAO, Xin-jun ZHOU, Bin HUANG, Gao-chao ZHANG, Ping HUANG, Ya-hui WANG, Kai-jun MA, Feng CHEN, Dong-chuan ZHANG, Ji ZHANG. Sex Estimation of Medial Aspect of the Ischiopubic Ramus in Adults Based on Deep Learning[J]. Journal of Forensic Medicine, 2023, 39(2): 129-136.

Figures/Tables 5

Fig. 1 The performance of the training verification and testing of the initialized learning models of each side of the MIPR images

Tab. 1 The performance of deep learning models for sex estimation in the testing dataset

模型	总准确率/ %	女性准确率/ %	男性准确率/ %	阳性预测值/ %	阴性预测值/ %	AUC值
模型1（右侧MIPR初始化学习模型）	95.7	95.7	95.7	95.7	95.7	0.989
模型2（左侧MIPR初始化学习模型）	92.1	88.6	95.7	95.4	89.3	0.967
模型3（双侧MIPR初始化学习模型）	94.6	92.1	97.1	97.0	92.5	0.983
模型4（双侧MIPR迁移学习模型）	95.7	95.7	95.7	95.7	95.7	0.975

Fig. 2 The performance of the training verification and testing of the initialized learning model trained with both sides of the MIPR images

Fig. 3 The performance of the training verification and testing of the transfer learning model trained with both sides of the MIPR images

Fig. 4 Heatmaps of the deep learning models of the MIPR

References 38

1	ZHANG Z， ONG S H， ZHONG X， et al. Efficient 3D dental identification via signed feature histogram and learning keypoint detection[J]. Pattern Recogn，2016，60：189-204. doi：10.1016/j.patcog.2016. 05.007 .
2	GÓMEZ O， IBÁÑEZ O， VALSECCHI A， et al. 3D-2D silhouette-based image registration for comparative radiography-based forensic identification[J]. Pattern Recogn，2018，83：469-480. doi：10.1016/j.pat cog.2018.06.011 .
3	HOLOBINKO A. Forensic human identification in the United States and Canada： A review of the law， admissible techniques， and the legal implications of their application in forensic cases[J]. Forensic Sci Int，2012，222（1/2/3）：394.e1-394.e13. doi：10 .
	1016/j.forsciint.2012.06.001.
4	LEUTENEGGER W. Functional aspects of pelvic morphology in Simian Primates[J]. J Hum Evol，1974，3（3）：207-222. doi：10.1016/0047-2484（74）90179-1 .
5	PHENICE T W. A newly developed visual method of sexing the os pubis[J]. Am J Phys Anthropol，1969，30（2）：297-301. doi：10.1002/ajpa.1330300214 .
6	KELLEY M A. Phenice’s visual sexing technique for the os pubis： A critique[J]. Am J Phys Anthropol，1978，48（1）：121-122. doi：10.1002/ajpa.1330480118 .
7	KLALES A R， OUSLEY S D， VOLLNER J M. A revised method of sexing the human innominate using Phenice’s nonmetric traits and statistical methods[J]. Am J Phys Anthropol，2012，149（1）：104-114. doi：10 .
	1002/ajpa.22102.
8	KLALES A R， COLE S J. Improving nonmetric sex classification for hispanic individuals[J]. J Forensic Sci，2017，62（4）：975-980. doi：10.1111/1556-4029.13391 .
9	JOHNSTONE-BELFORD E， FLAVEL A， FRANKLIN D. Morphoscopic observations in clinical pelvic MDCT scans： Assessing the accuracy of the Phenice traits for sex estimation in a Western Australian population[J]. J Forensic Radiol Im，2018，12：5-10. doi：10 .
	1016/j.jofri.2018.02.003.
10	GÓMEZ-VALDÉS J A， MENÉNDEZ GARMENDIA A， GARCÍA-BARZOLA L， et al. Recalibration of the Klales et al. （2012） method of sexing the human innominate for Mexican populations[J]. Am J Phys Anthropol，2017，162（3）：600-604. doi：10.1002/ajpa.23157 .
11	SUTHERLAND L D， SUCHEY J M. Use of the ventral arc in pubic sex determination[J]. J Forensic Sci，1991，36（2）：501-511.
12	LOVELL N C. Test of Phenice’s technique for determining sex from the os pubis[J]. Am J Phys Anthropol，1989，79（1）：117-120. doi：10.1002/ajpa.1330 790112 .
13	MCFADDEN C， OXENHAM M F. Revisiting the Phenice technique sex classification results reported by MacLaughlin and Bruce （1990）[J]. Am J Phys Anthropol，2016，159（1）：182-183. doi：10.1002/ajpa. 22839 .
14	LESCIOTTO K M， DOERSHUK L J . Accuracy and reliability of the Klales et al. （2012） morphoscopic pelvic sexing method[J]. J Forensic Sci，2018，63（1）：214-220. doi：10.1111/1556-4029.13501 .
15	ZHAN M J， FAN F， QIU L R， et al. Estimation of stature and sex from sacrum and coccyx measurements by multidetector computed tomography in Chinese[J]. Leg Med （Tokyo），2018，34：21-26. doi：10.1016/j.legalmed.2018.07.003 .
16	陶建华，陈聪，张怀宇，等. 鼻骨区骨折深度学习模型的建立和临床效能评估[J].放射学实践，2021，36（8）：959-964. doi：10.13609/j.cnki.1000-0313.2021.08.003 .
	TAO J H， CHEN C， ZHANG H Y， et al. Establishment and evaluation of the clinical efficacy of deep learning model for detection of nasal bone fracture on CT images[J]. Fangshexue Shijian，2021，36（8）：959-964.
17	钟碧霞，周冠群，许文琪，等. 基于深度学习的结肠癌病理图片分类研究[J].中国卫生统计，2021，38（3）：363-367. doi：10.3969/j.issn.1002-3674.2021.03.010 .
	ZHONG B X， ZHOU G Q， XU W Q， et al. Classification for pathological images of colon cancer with deep learning[J]. Zhongguo Weisheng Tongji，2021，38（3）：363-367.
18	刘晶，鲜军舫，李书玲，等. 深度学习模型检测胸部CT肺结节的临床效能评估[J].实用放射学杂志，2021，37（5）：732-735，767. doi：10.3969/j.issn.1002-1671.2021.05.010 .
	LIU J， XIAN J F， LI S L， et al. Evaluation of the efficacy of deep learning model in detecting pulmonary nodules on chest CT images[J]. Shiyong Fangshexue Zazhi，2021，37（5）：732-735，767.
19	ESTEVA A， KUPREL B， NOVOA R A， et al. Dermatologist-level classification of skin cancer with deep neural networks[J]. Nature，2017，542（7639）：115-118. doi：10.1038/nature21056 .
20	SPAMPINATO C， PALAZZO S， GIORDANO D， et al. Deep learning for automated skeletal bone age assessment in X-ray images[J]. Med Image Anal，2017，36：41-51. doi：10.1016/j.media.2016.10.010 .
21	ZHOU Y， ZHANG J， HUANG J， et al. Digital whole-slide image analysis for automated diatom test in forensic cases of drowning using a convolutional neural network algorithm[J]. Forensic Sci Int，2019，302：109922. doi：10.1016/j.forsciint.2019.109922 .
22	ZHANG J， ZHOU Y， VIEIRA D N， et al. An efficient method for building a database of diatom populations for drowning site inference using a deep learning algorithm[J]. Int J Legal Med，2021，135（3）：817-827. doi：10.1007/s00414-020-02497-5 .
23	CAO Y， MA Y， VIEIRA D N， et al. A potential method for sex estimation of human skeletons using deep learning and three-dimensional surface scanning[J]. Int J Legal Med，2021，135（6）：2409-2421. doi：10.1007/s00414-021-02675-z .
24	LECUN Y， BENGIO Y， HINTON G. Deep learning[J]. Nature，2015，521（7553）：436-444. doi：10.1038/ nature14539 .
25	SZEGEDY C， IOFFE S， VANHOUCKE V， et al. Inception-v4， Inception-ResNet and the impact of residual connections on learning[C]// Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence. San Francisco， California， USA，2017：4278-4284.
26	胡婷鸿，火忠，刘太昂，等. 基于深度学习实现维吾尔族青少年左手腕关节骨龄自动化评估[J].法医学杂志，2018，34（1）：27-32. doi：10.3969/j.issn.1004-5619.2018.01.006 .
	HU T H， HUO Z， LIU T A， et al. Automated assessment for bone age of left wrist joint in Uyghur teenagers by deep learning[J]. Fayixue Zazhi，2018，34（1）：27-32.
27	彭丽琴，万雷，汪茂文，等. 运用3种卷积神经网络模型对青少年骨盆骨龄评估的比较[J].法医学杂志，2020，36（5）：622-630. doi：10.12116/j.issn.1004-5619.2020.05.004 .
	PENG L Q， WAN L， WANG M W， et al. Comparison of three CNN models applied in bone age assessment of pelvic radiographs of adolescents[J]. Fayixue Zazhi，2020，36（5）：622-630.
28	PENG L Q， GUO Y C， WAN L， et al. Forensic bone age estimation of adolescent pelvis X-rays based on two-stage convolutional neural network[J]. Int J Legal Med，2022，136（3）：797-810. doi：10.1007/s004 14-021-02746-1 .
29	KLALES A R， VOLLNER J M， OUSLEY S D. A new metric procedure for the estimation of sex and ancestry from the human innominate[Z/OL]. [2022-04-12]. .
30	VOLLNER J M， KLALES A R， OUSLEY S D. Sexing of the human innominate using non-metric traits and statistical analysis[J/OL]. [2022-04-12]. .
31	COURTIOL P， MAUSSION C， MOARII M， et al. Deep learning-based classification of mesothelioma improves prediction of patient outcome[J]. Nat Med，2019，25（10）：1519-1525. doi：10.1038/s41591-019-05 83-3 .
32	GULSHAN V， PENG L， CORAM M， et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs[J]. JAMA，2016，316（22）：2402-2410. doi：10.1001/jama.2016.17216 .
33	DUNNMON J A， YI D， LANGLOTZ C P， et al. Assessment of convolutional neural networks for automated classification of chest radiographs[J]. Radio-logy，2019，290（2）：537-544. doi：10.1148/radiol.20181 81422 .
34	KERMANY D S， GOLDBAUM M， CAI W， et al. Identifying medical diagnoses and treatable diseases by image-based deep learning[J]. Cell，2018，172（5）：1122-1131.e9. doi：10.1016/j.cell.2018.02.010 .
35	COUDRAY N， OCAMPO P S， SAKELLAROPOU-LOS T， et al. Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning[J]. Nat Med，2018，24（10）：1559-1567. doi：10.1038/s41591-018-0177-5 .
36	KIM H E， COSA-LINAN A， SANTHANAM N， et al. Transfer learning for medical image classification： A literature review[J]. BMC Med Imaging，2022，22（1）：69. doi：10.1186/s12880-022-00793-7 .
37	OSIPENKO S， BOTASHEV K， NIKOLAEV E， et al. Transfer learning for small molecule retention predictions[J]. J Chromatogr A，2021，1644：462119. doi：10.1016/j.chroma.2021.462119 .
38	JAIN S， SINGHANIA U， TRIPATHY B， et al. Deep learning-based transfer learning for classification of skin cancer[J]. Sensors （Basel），2021，21（23）：8142. doi：10.3390/s21238142 .

[1]	Jia-xuan HAN, Shi-hui SHEN, Yi-wen WU, Xiao-dan SUN, Tian-nan CHEN, Jiang TAO. Adolescents and Children Age Estimation Using Machine Learning Based on Pulp and Tooth Volumes on CBCT Images [J]. Journal of Forensic Medicine, 2024, 40(2): 143-148.
[2]	Lei SHI, Ye XUE, Li-rong QIU, Ting LU, Fei FAN, Yu-chi ZHOU, Zhen-hua DENG. Research Progress on Dental Age Estimation Based on MRI Technology [J]. Journal of Forensic Medicine, 2024, 40(2): 112-117.
[3]	Xuan WEI, Yu-shan CHEN, Jie DING, Chang-xing SONG, Jun-jing WANG, Zhao PENG, Zhen-hua DENG, Xu YI, Fei FAN. Age Estimation by Machine Learning and CT-Multiplanar Reformation of Cranial Sutures in Northern Chinese Han Adults [J]. Journal of Forensic Medicine, 2024, 40(2): 128-134.
[4]	Yu-xin GUO, Wen-qing BU, Yu TANG, Di WU, Hui YANG, Hao-tian MENG, Yu-cheng GUO. Dental Age Estimation in Northern Chinese Han Children and Adolescents Using Demirjian’s Method Combined with Machine Learning Algorithms [J]. Journal of Forensic Medicine, 2024, 40(2): 135-142.
[5]	Dan-yang LI, Yu PAN, Hui-ming ZHOU, Lei WAN, Cheng-tao LI, Mao-wen WANG, Ya-hui WANG. Application of Medical Statistical and Machine Learning Methods in the Age Estimation of Living Individuals [J]. Journal of Forensic Medicine, 2024, 40(2): 118-127.
[6]	Huai-han ZHANG, Yong-jie CAO, Ji ZHANG, Jian XIONG, Ji-wei MA, Xiao-tong YANG, Ping HUANG, Yong-gang MA. Adults Ischium Age Estimation Based on Deep Learning and 3D CT Reconstruction [J]. Journal of Forensic Medicine, 2024, 40(2): 154-163.
[7]	De-min HUO, Kai-jun MA, Jing-lan XU, Xu SONG, Xiao-yan MAO, Xia LIU, Kai-fang ZHAO, Jian ZHANG, Meng DU. Determining Whether an Individual is 18 Years or Older Based on the Third Molar Root Pulp Visibility in East China [J]. Journal of Forensic Medicine, 2024, 40(2): 149-153.
[8]	Hong-xia HAO, Ya-hui WANG, Zhi-lu ZHOU, Tai-ang LIU, Jin CHEN, Yu-heng HE, Lei WAN, Wen-tao XIA. Research Progress of Age Estimation in the Living by Knee Joint MRI [J]. Journal of Forensic Medicine, 2023, 39(1): 66-71.
[9]	Xiao-tong YANG, Cheng-hui SUN, Yong-gang MA, Yong-jie CAO, Jian XIONG, Ji ZHANG, Ping HUANG. Sex Estimation of Han Adults in Western China Based on Three-Dimensional Cranial CT Reconstruction [J]. Journal of Forensic Medicine, 2023, 39(1): 27-33.
[10]	Jing-jing MA, Min SHANG, Da-wei QIN, Wen-zhao REN, Jia-yin SHI, Xiao-yun ZHAO, Ze-wei YAO, Fan LI. Relationship between the Height of Alveolar Bone Resorption and Sex and Age in Adolescents [J]. Journal of Forensic Medicine, 2022, 38(5): 589-594.
[11]	Dong-chuan ZHANG, Geng FEI, Ting-ting CHEN, Lu-yi XU, De-lun YU, Tian-ye ZHANG. Comparison of Three Methods for Skeleton Age Estimation [J]. Journal of Forensic Medicine, 2022, 38(3): 319-323.
[12]	Fei-xiang WANG, Rui JI, Lu-ming ZHANG, Peng WANG, Tai-ang LIU, Lu-jie SONG, Mao-wen WANG, Zhi-lu ZHOU, Hong-xia HAO, Wen-tao XIA. Pelvic Injury Discriminative Model Based on Data Mining Algorithm [J]. Journal of Forensic Medicine, 2022, 38(3): 350-354.
[13]	Qi-fan YANG, Xue-yang SUN, Yan-bin WANG, Zhi-ling TIAN, He-wen DONG, Lei WAN, Dong-hua ZOU, Xiao-tian YU, Guang-zheng ZHANG, Ning-guo LIU. Automatic Identification of Brain Injury Mechanism Based on Deep Learning [J]. Journal of Forensic Medicine, 2022, 38(2): 223-230.
[14]	Yong-zheng ZHU, Ji ZHANG, Qi CHENG, Hui-xiao YU, Kai-fei DENG, Jian-hua ZHANG, Zhi-qiang QIN, Jian ZHAO, Jun-hong SUN, Ping HUANG. Comparison among Four Deep Learning Image Classification Algorithms in AI-based Diatom Test [J]. Journal of Forensic Medicine, 2022, 38(1): 31-39.
[15]	Yong-zheng ZHU, Ji ZHANG, Qi CHENG, Kai-fei DENG, Kai-jun MA, Jian-hua ZHANG, Jian ZHAO, Jun-hong SUN, Ping HUANG, Zhi-qiang QIN. Research Progress of Automatic Diatom Test by Artificial Intelligence [J]. Journal of Forensic Medicine, 2022, 38(1): 14-19.