Adults Ischium Age Estimation Based on Deep Learning and 3D CT Reconstruction

doi:10.12116/j.issn.1004-5619.2023.231003

Abstract

Abstract:

Objective To develop a deep learning model for automated age estimation based on 3D CT reconstructed images of Han population in western China, and evaluate its feasibility and reliability. Methods The retrospective pelvic CT imaging data of 1 200 samples （600 males and 600 females） aged 20.0 to 80.0 years in western China were collected and reconstructed into 3D virtual bone models. The images of the ischial tuberosity feature region were extracted to create sex-specific and left/right site-specific sample libraries. Using the ResNet34 model, 500 samples of different sexes were randomly selected as training and verification set, the remaining samples were used as testing set. Initialization and transfer learning were used to train images that distinguish sex and left/right site. Mean absolute error （MAE） and root mean square error （RMSE） were used as primary indicators to evaluate the model. Results Prediction results varied between sexes, with bilateral models outperformed left/right unilateral ones, and transfer learning models showed superior performance over initial models. In the prediction results of bilateral transfer learning models, the male MAE was 7.74 years and RMSE was 9.73 years, the female MAE was 6.27 years and RMSE was 7.82 years, and the mixed sexes MAE was 6.64 years and RMSE was 8.43 years. Conclusion The skeletal age estimation model, utilizing ischial tuberosity images of Han population in western China and employing the ResNet34 combined with transfer learning, can effectively estimate adult ischium age.

Key words: forensic anthropology, age estimation, deep learning, three-dimensional reconstruction, pelvis, ischial tuberosity, transfer learning, Han population

CLC Number:

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.

Figures/Tables 7

References 38

1	占梦军，张世杰，陈虎，等. 骨龄自动化评估的研究进展[J].法医学杂志，2020，36（2）：249-255. doi：10.12116/j.issn.1004-5619.2020.02.019 .
	ZHAN M J， ZHANG S J， CHEN H， et al. Research progress on automatic assessment of bone age[J]. Fa-yixue Zazhi，2020，36（2）：249-255.
2	GARVIN H M， PASSALACQUA N V. Current practices by forensic anthropologists in adult skeletal age estimation[J]. J Forensic Sci，2012，57（2）：427-433. doi：10.1111/j.1556-4029.2011.01979.x .
3	GARVIN H M， PASSALACQUA N V， UHL N M， et al. Developments in forensic anthropology： Age-at-death estimation[M]// DIRKMAAT D. A Companion to forensic anthropology， Malden， Massachusetts： Wiley-Blackwell，2012：202-223.
4	MÁRQUEZ-GRANT N. An overview of age estimation in forensic anthropology： Perspectives and practical considerations[J]. Ann Hum Biol，2015，42（4）：308-322. doi：10.3109/03014460.2015.1048288 .
5	裘诗文，涂梦，范飞，等. 应用颅缝薄层CT扫描推断汉族成人年龄[J].法医学杂志，2020，36（4）：507-513. doi：10.12116/j.issn.1004-5619.2020.04.015 .
	QIU S W， TU M， FAN F， et al. Age estimation in Han adults by thin-layer CT scan of cranial sutures[J]. Fayixue Zazhi，2020，36（4）：507-513.
6	GUO Y C， WANG Y H， OLZE A， et al. Dental age estimation based on the radiographic visibility of the periodontal ligament in the lower third molars： Application of a new stage classification[J]. Int J Legal Med，2020，134（1）：369-374. doi：10.1007/s00414-019-02178-y .
7	LU T， DIAO Y R， TANG X E， et al. Deep learning enables automatic adult age estimation based on CT reconstruction images of the costal cartilage[J]. Eur Radiol，2023，33（11）：7519-7529. doi：10.1007/s00330-023-09761-3 .
8	熊剪，曹永杰，马永刚，等. 基于机器学习和CT三维重建技术的成人耻骨联合年龄推断[J].复旦学报（医学版），2023，50（5）：731-742. doi：10.3969/j.issn.1672-8467.2023.05.012 .
	XIONG J， CAO Y J， MA Y G， et al. A method based on CT three-dimensional reconstruction and machine learning for age estimation of pubic symphysis in adults[J]. Fudan Xuebao （Medical sciences），2023，50（5）：731-742.
9	LISTI G A. The use of entheseal changes in the femur and os coxa for age assessment[J]. J Forensic Sci，2016，61（1）：12-18. doi：10.1111/1556-4029.12905 .
10	ZHANG K， DONG X A， CHEN X G， et al. The ossification of the ischial tuberosity for forensic age diagnostics in conventional radiography[J]. Aust J Forensic Sci，2014，46（4）：455-462. doi：10.1080/00450618.2014.891652 .
11	SCHEUER L. Application of osteology to forensic medicine[J]. Clin Anat，2002，15（4）：297-312. doi：10 .
	1002/ca.10028.
12	STEYN M， BECKER P J， L’ABBÉ E N， et al. An assessment of the repeatability of pubic and ischial measurements[J]. Forensic Sci Int，2012，214（1/2/3）：210.e1-210.e4. doi：10.1016/j.forsciint.2011.07.049 .
13	RISSECH C， GARCÍA M， MALGOSA A. Sex and age diagnosis by ischium morphometric analysis[J]. Forensic Sci Int，2003，135（3）：188-196. doi：10.1016/s0379-0738（03）00215-9 .
14	TORIMITSU S， MAKINO Y， SAITOH H， et al. Stature estimation in Japanese cadavers based on pelvic measurements in three-dimensional multidetector computed tomographic images[J]. Int J Legal Med，2015，129（3）：633-639. doi：10.1007/s00414-014-1000-6 .
15	IMAIZUMI K， USUI S， TANIGUCHI K， et al. Development of an age estimation method for bones based on machine learning using post-mortem computed tomography images of bones[J]. Forensic Imag，2021，26：200477. doi：10.1016/j.fri.2021.200477 .
16	马永刚，曹永杰，赵益花，等. 基于深度学习实现成人坐骨耻骨支内侧缘的性别推断[J].法医学杂志，2023，39（2）：129-136，143. doi：10.12116/j.issn.1004-5619.2022.220505 .
	MA Y G， CAO Y J， ZHAO Y H， et al. Sex estimation of medial aspect of the ischiopubic ramus in adults based on deep learning[J]. Fayixue Zazhi，2023，39（2）：129-136，143.
17	XIONG J， MA Y， CAO Y， et al. Age estimation by modified Suchey-Brooks method using three-dimensional reconstructed CT images of Chinese Han population[J]. Leg Med （Tokyo），2023，65：102304. doi：10.1016/j.legalmed.2023.102304 .
18	杨孝通，孙丞辉，马永刚，等. 基于CT三维重建技术推断西部地区汉族成人颅骨性别[J].法医学杂志，2023，39（1）：27-33. doi：10.12116/j.issn.1004-5619.2022.220101 .
	YANG X T， SUN C H， MA Y G， et al. Sex estimation of Han adults in western China based on three-dimensional cranial CT reconstruction[J]. Fayi-xue Zazhi，2023，39（1）：27-33.
19	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 .
20	LECUN Y， BENGIO Y， HINTON G. Deep lear-ning[J]. Nature，2015，521（7553）：436-444. doi：10.1038/nature14539 .
21	HE K， ZHANG X， REN S， et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）， Nevada， USA： Las Vegas，2016：27-30.
22	PAN S J， YANG Q. A survey on transfer learning[J]. IEEE Trans Knowl Data Eng，2010，22（10）：1345-1359. doi：10.1109/TKDE.2009.191 .
23	RODRIGUEZ J D， PEREZ A， LOZANO J A. Sensitivity analysis of k-fold cross validation in prediction error estimation[J]. IEEE Trans Pattern Anal Mach Intell，2010，32（3）：569-575. doi：10.1109/TPAMI.2009.187 .
24	CALCE S E， ROGERS T L. Evaluation of age estimation technique： Testing traits of the acetabulum to estimate age at death in adult males[J]. J Forensic Sci，2011，56（2）：302-311. doi：10.1111/j.1556-4029. 2011.01700.x .
25	KOLESOVA O， KOLESOVS A， VETRA J. Age-related trends of lesser pelvic architecture in females and males： A computed tomography pelvime-try study[J]. Anat Cell Biol，2017，50（4）：265-274. doi：10.5115/acb.2017.50.4.265 .
26	MALLARD A M， SAVELL K R R， AUERBACH B M. Morphological integration of the human pelvis with respect to age and sex[J]. Anat Rec （Hoboken），2017，300（4）：666-674. doi：10.1002/ar.23547 .
27	KONDOU H， MOROHASHI R， ICHIOKA H， et al. Deep neural networks-based age estimation of cadavers using CT imaging of vertebrae[J]. Int J Environ Res Public Health，2023，20（6）：4806. doi：10.3390/ijerph20064806 .
28	SAN-MILLÁN M， RISSECH C， TURBÓN D. New approach to age estimation of male and female adult skeletons based on the morphological characteristics of the acetabulum[J]. Int J Legal Med，2017，131（2）：501-525. doi：10.1007/s00414-016-1406-4 .
29	MAYS S. The effect of factors other than age upon skeletal age indicators in the adult[J]. Ann Hum Biol，2015，42（4）：332-341. doi：10.3109/03014460.2015.1044470 .
30	MERRITT C E. The influence of body size on adult skeletal age estimation methods[J]. Am J Phys Anthropol，2015，156（1）：35-57. doi：10.1002/ajpa.22626 .
31	LI Y， HUANG Z， DONG X， et al. Forensic age estimation for pelvic X-ray images using deep learning[J]. Eur Radiol，2019，29（5）：2322-2329. doi：10.1007/s00330-018-5791-6 .
32	SHIN H C， ROTH H R， GAO M C， et al. Deep convolutional neural networks for computer-aided detection： CNN architectures， dataset characteristics and transfer learning[J]. IEEE Trans Med Imaging，2016，35（5）：1285-1298. doi：10.1109/TMI.2016.2528162 .
33	TAJBAKHSH N， SHIN J Y， GURUDU S R， et al. Convolutional neural networks for medical image analysis： Full training or fine tuning?[J]. IEEE Trans Med Imaging，2016，35（5）：1299-1312. doi：10.1109/TMI.2016.2535302 .
34	NAGATA T， TANIGUCHI K， OGAWA Y， et al. Age estimation from human skeleton by machine learning[C]// 2020 21st International Conference on Research and Education in Mechatronics （REM）. Cracow， Poland，2020：1-4. doi：10.1109/REM49740.2020.9313896 .
35	田雪梅，张继宗，闵建雄，等. 青少年骨关节X线片的骨龄研究[J].刑事技术，2001（2）：6-11.
	TIAN X M， ZHANG J Z， MIN J X， et al. Legal age estimation of northern Chinese by skeletal maturity[J]. Xingshi Jishu，2001（2）：6-11.
36	朱广友，王亚辉，万雷. 中国青少年骨龄鉴定标准图谱法[M].上海：上海科学技术文献出版社，2016：61-81.
	ZHU G Y， WANG Y H， WAN L. Determination of skeleton age for Chinese teenagers by standard radiographic atlas[M]. Shanghai： Shanghai Scientific and Technological Literature Press，2016：61-81.
37	LOVEJOY C O， MEINDL R S， PRYZBECK T R， et al. Chronological metamorphosis of the auricular surface of the ilium： A new method for the determination of adult skeletal age at death[J]. Am J Phys Anthropol，1985，68（1）：15-28. doi：10.1002/ajpa.1330680103 .
38	MILES A E W， BULMAN J S. Growth curves of immature bones from a Scottish island population of sixteenth to mid-nineteenth century： Shoulder girdle， ilium， pubis and ischium[J]. Int J Osteoarchaeol，1995，5（1）：15-27. doi：10.1002/oa.1390050103 .

模型	性别	部位	MAE/岁	RMSE/岁	MBE/岁	AE<5%	AE<10%
初始模型	男性	左侧	7.76/9.13	10.12/11.27	-0.16/-1.52	40.00%/36.00%	71.00%/56.00%
		右侧	7.88/9.18	10.53/11.62	-1.26/-3.25	43.60%/37.00%	71.80%/57.00%
		双侧	7.58/8.59	9.90/10.90	-0.62/-2.02	44.00%/40.00%	72.30%/63.50%
	女性	左侧	7.24/7.49	9.22/9.24	-0.12/0.97	45.00%/41.00%	73.80%/70.00%
		右侧	7.60/7.71	9.58/9.47	-0.99/-1.92	42.80%/39.00%	73.80%/69.00%
		双侧	6.84/7.40	8.70/9.02	-1.07/-1.55	46.60%/37.50%	74.60%/71.50%
	混合性别	左侧	7.48/8.17	9.58/10.11	-0.92/-1.58	43.90%/35.50%	72.20%/66.50%
		右侧	7.59/7.75	9.56/9.77	-0.39/-1.55	40.90%/40.50%	72.00%/70.00%
		双侧	7.17/7.67	9.21/9.69	-0.82/-1.61	45.30%/42.00%	73.70%/69.50%
迁移模型	男性	左侧	6.72/8.28	8.85/10.41	-0.50/-1.99	49.40%/42.00%	79.20%/60.00%
		右侧	6.91/7.80	8.99/9.86	-1.11/-2.28	46.80%/40.00%	75.40%/71.00%
		双侧	6.26/7.74	8.14/9.73	-0.42/-1.74	51.10%/39.00%	78.80%/67.00%
	女性	左侧	5.99/6.62	7.72/7.94	-0.63/0.21	51.40%/42.00%	82.00%/78.00%
		右侧	5.92/6.37	7.51/8.21	0.29/0.02	50.20%/50.00%	81.60%/77.00%
		双侧	5.60/6.27	7.19/7.82	-0.26/-0.11	55.00%/45.50%	83.50%/81.00%
	混合性别	左侧	6.19/7.26	8.05/9.05	-0.68/-1.13	51.60%/40.50%	80.50%/74.00%
		右侧	6.18/6.74	8.03/8.72	-0.41/-0.91	51.00%/47.50%	79.40%/76.00%
		双侧	6.04/6.64	7.85/8.43	-0.67/-1.09	53.10%/46.00%	80.90%/77.75%

模型	性别	部位	MAE/岁	RMSE/岁	MBE/岁	AE<5%	AE<10%
初始模型	男性	左侧	7.76/9.13	10.12/11.27	-0.16/-1.52	40.00%/36.00%	71.00%/56.00%
		右侧	7.88/9.18	10.53/11.62	-1.26/-3.25	43.60%/37.00%	71.80%/57.00%
		双侧	7.58/8.59	9.90/10.90	-0.62/-2.02	44.00%/40.00%	72.30%/63.50%
	女性	左侧	7.24/7.49	9.22/9.24	-0.12/0.97	45.00%/41.00%	73.80%/70.00%
		右侧	7.60/7.71	9.58/9.47	-0.99/-1.92	42.80%/39.00%	73.80%/69.00%
		双侧	6.84/7.40	8.70/9.02	-1.07/-1.55	46.60%/37.50%	74.60%/71.50%
	混合性别	左侧	7.48/8.17	9.58/10.11	-0.92/-1.58	43.90%/35.50%	72.20%/66.50%
		右侧	7.59/7.75	9.56/9.77	-0.39/-1.55	40.90%/40.50%	72.00%/70.00%
		双侧	7.17/7.67	9.21/9.69	-0.82/-1.61	45.30%/42.00%	73.70%/69.50%
迁移模型	男性	左侧	6.72/8.28	8.85/10.41	-0.50/-1.99	49.40%/42.00%	79.20%/60.00%
		右侧	6.91/7.80	8.99/9.86	-1.11/-2.28	46.80%/40.00%	75.40%/71.00%
		双侧	6.26/7.74	8.14/9.73	-0.42/-1.74	51.10%/39.00%	78.80%/67.00%
	女性	左侧	5.99/6.62	7.72/7.94	-0.63/0.21	51.40%/42.00%	82.00%/78.00%
		右侧	5.92/6.37	7.51/8.21	0.29/0.02	50.20%/50.00%	81.60%/77.00%
		双侧	5.60/6.27	7.19/7.82	-0.26/-0.11	55.00%/45.50%	83.50%/81.00%
	混合性别	左侧	6.19/7.26	8.05/9.05	-0.68/-1.13	51.60%/40.50%	80.50%/74.00%
		右侧	6.18/6.74	8.03/8.72	-0.41/-0.91	51.00%/47.50%	79.40%/76.00%
		双侧	6.04/6.64	7.85/8.43	-0.67/-1.09	53.10%/46.00%	80.90%/77.75%

年龄段/岁	男性	女性	混合性别
20.0~<30.0	6.41/6.01	4.79/4.69	4.59/4.36
30.0~<40.0	5.64/1.97	4.08/2.54	4.85/2.09
40.0~<50.0	6.29/1.60	5.51/2.80	6.04/2.09
50.0~<60.0	7.41/0.26	4.94/0.93	5.95/0.42
60.0~<70.0	8.34/-7.81	8.52/-5.97	8.56/-6.88
70.0~80.0	11.97/-11.28	9.96/-5.69	9.67/-8.89

年龄段/岁	男性	女性	混合性别
20.0~<30.0	6.41/6.01	4.79/4.69	4.59/4.36
30.0~<40.0	5.64/1.97	4.08/2.54	4.85/2.09
40.0~<50.0	6.29/1.60	5.51/2.80	6.04/2.09
50.0~<60.0	7.41/0.26	4.94/0.93	5.95/0.42
60.0~<70.0	8.34/-7.81	8.52/-5.97	8.56/-6.88
70.0~80.0	11.97/-11.28	9.96/-5.69	9.67/-8.89

[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]	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.
[7]	Yi-hang HUANG, Wei-bo LIANG, Hui JIAN, Sheng-qiu QU. Modeling Methods and Influencing Factors for Age Estimation Based on DNA Methylation [J]. Journal of Forensic Medicine, 2023, 39(6): 601-607.
[8]	Xing-yu MA, Hao CHENG, Zhong-duo ZHANG, Ye-ming LI, Dong ZHAO. Research Progress of Metabolomics Techniques Combined with Machine Learning Algorithm in Wound Age Estimation [J]. Journal of Forensic Medicine, 2023, 39(6): 596-600.
[9]	Zi-wei WANG, Cheng-tao LI, Xi-ling LIU. Cross-Platform Application of DNA Methylation Age Estimation Model in Eastern Chinese Han Population [J]. Journal of Forensic Medicine, 2023, 39(5): 441-446.
[10]	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.
[11]	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.
[12]	Zi-wei WANG, Qian-nan XU, Cheng-tao LI, Xi-ling LIU. Age Estimation Based on DNA Methylation and Its Application Prospects in Forensic Medicine [J]. Journal of Forensic Medicine, 2023, 39(1): 72-82.
[13]	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.
[14]	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.
[15]	Yan-ru XI, Yuan-xin LIU, Na FENG, Zhen GU, Jun-hong SUN, Jie CAO, Qian-qian JIN, Qiu-xiang DU. Application of Linear Regression Model of Gpnmb Gene in Rat Injury Time Estimation [J]. Journal of Forensic Medicine, 2022, 38(4): 468-472.