Top

Published in:

01-11-2021 | Artificial Intelligence | Short Communication

Analysis of the performance of machine learning and deep learning methods for sex estimation of infant individuals from the analysis of 2D images of the ilium

Authors: Raúl Fernández Ortega, Javier Irurita, Enrique José Estévez Campo, Pablo Mesejo

Published in: International Journal of Legal Medicine | Issue 6/2021

Abstract

Reducing the subjectivity of the methods used for biological profile estimation is, at present, a priority research line in forensic anthropology. To achieve this, artificial intelligence (AI) techniques can be a valuable tool yet to be exploited in this discipline. The goal of this study is to compare the effectiveness of different machine learning (ML) methods with the visual assessment of an expert to estimate the sex of infant skeletons from images of the ilium. Photographs of the ilium of 135 individuals, age between 5 months of gestation and 6 years, from the collection of identified infant skeletons of the University of Granada have been used, and classic ML and deep learning (DL) techniques have been applied to develop prediction algorithms. To assess their effectiveness, the results have been compared with those obtained by a forensic expert, who has estimated the sex from each photograph through direct observation and subjective assessment following the criteria described by Schutkowsky in 1993. The results show that the algorithms obtained using DL techniques offer an accuracy of 59%, very close to the 61% obtained by the expert, and 10 percentual points better than classic ML techniques. This study offers promising results and represents the first AI-based approach for estimating sex in infant individuals using photographs of the ilium.

https://keras.io/

https://scikit-learn.org/

Brooks S, Suchey J (1990) Skeletal age determination based on the os pubis: a comparison of the Acáfidi-Nemeskéri and Suchey-Brooks methods. Hum Evol 5(3):227–238CrossRef

Buckberry J, Chamberlain A (2002) Age estimation from the auricular surface of the ilium: a revised method. Am J Phys Anthropol 119:231–239PubMedCrossRef

Iscan M, Loth S, Wright R (1984) Metamorphosis at the sternal rib end: a new method to estimate age at death in white males. Am J Phys Anthropol 65:147–156PubMedCrossRef

Kellinghaus M, Schulz R, Vieth V, Schmidt S, Schmeling A (2010) Forensic age estimation in living subjects based on the ossification status of the medial clavicular epiphysis as revealed by thin-slice multidetector computed tomography. Int J Legal Med 124:149–154PubMedCrossRef

Demirjian A, Goldstein H, Tanner J (1973) A new system of dental age assessment. Hum Biol 45:211–227PubMed

Buikstra J, Ubelaker D (1994) Standards for data collection from human skeletal remains: Proceedings of a seminar at the field museum of natural history. Arkansas Archeological Survey Series, Fayetteville, Arkansas

Lesciotto K (2015) The impact of Daubert on the admissibility of forensic anthropology expert testimony. J Foren Sci 60(3):549–555PubMedCrossRef

Alpaydin E (2004) Introduction to machine learning. MIT Press, Cambridge, Massachusetts

Bishop C (2006) Pattern recognition and machine learning, Springer

10.

Jordan M (2015) Machine learning: trends, perspectives, and prospects. Science 349(6245):255–60PubMedCrossRef

11.

LeCun Y et al (2015) Deep learning. Natura 521(7553):436–444CrossRef

12.

Schmidhuber J (2015) Deep learning in neural networks: an overview. Neural netw 61:85–117PubMedCrossRef

13.

Goodfellow I, Bengio Y, Courville A (2016) Deep learning (adaptive computation and machine learning series). MIT Press. https://www.deeplearningbook.org

14.

Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg A, Li F (2015) Imagenet large scale visual recognition challenge. Int J Comput Vis 115(3):211–252CrossRef

15.

Shi S, Wang Q, Xu P, Chu X (2016) Benchmarking state-of-the-art deep learning software tools. 2016 7th International Conference on Cloud Computing and Big Data (CCBD), pp 99–104. https://doi.org/10.1109/CCBD.2016.029

16.

Srivastasa N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958

17.

Hinton G, Deng L, Yu D, Dahl G, Mohamed A, Jaitly N, Senior A, Vanhoucke V, Nguyen P, Sainath T, Kingsbury B (2012) Deep neural networks for acoustic modeling in speech recognition. IEEE Signal Procg Mag 29:82–97CrossRef

18.

Krizhevsky A, Sutskever I, Hinton G (2012) ImageNet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 25:1090–1098

19.

Ciregan D, Meier U (2015) Multi-column deep neural networks for offline handwritten Chinese character classification. 2015 International Joint Conference on Neural Networks (IJCNN), pp 1–6. https://doi.org/10.1109/IJCNN.2015.7280516

20.

Silver D et al (2016) Mastering the game of Go with deep neural networks and tree search. Nature 529(7587):484–489PubMedCrossRef

21.

Girshick R, Donahue J, Darrell T, Malik J (2014) Rich feature hierarchies for accurate object detection and semantic segmentation. 2014 IEEE Conference on computer vision and pattern recognition, pp 580–587. https://doi.org/10.1109/CVPR.2014.81

22.

Taigman Y, Yang M, Ranzato M, Wolf L (2014) DeepFace: closing the gap to human-level performance in face verification. 2014 IEEE conference on computer vision and pattern recognition, pp 1701–1708. https://doi.org/10.1109/CVPR.2014.220.587

23.

Lathuilière S, Mesejo P, Alameda-Pineda X, Horaud R (2020) A comprehensive analysis of deep regression. IEEE Trans Pattern Anal Mach Intell 42(9):2065–2081PubMedCrossRef

24.

Lathuilière S, Juge R, Mesejo P, Muñoz-Salinas R, Horaud R (2017) Deep mixture of linear inverse regressions applied to head-pose estimation. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp 7149–7157. https://doi.org/10.1109/CVPR.2017.756

25.

Greenspan H, Ginneken B, Summers R (2016) Guest editorial deep learning in medical imaging: overview and future promise of an exciting new technique. IEEE Trans Med Imaging 35(5):1153–1159CrossRef

26.

Bewes J, Low A, Morphett A, Pate F, Henneberg M (2019) Artificial intelligence for sex determination of skeletal remains: application of a deep learning artificial neural network to human skulls. J Forensic Leg Med 62:40–43PubMedCrossRef

27.

Yang W, Liu X, Wang K, Hu J, Geng G, Feng J (2019) Sex determination of three-dimensional skull based on improved backpropagation neural network. Comput Math Methods Med (6):1–8

28.

Schutkowski H (1993) Sex determination of infant and juvenile skeletons: I. morphognostic features. Am J Phys Anthropol 90:199–205PubMedCrossRef

29.

Lewis M (2007) The bioarchaeology of children, 2nd edn. Cambridge University Press, Cambridge

30.

Sutter R (2003) Nonmetric subadult skeletal sexing traits: I. A blind test of the accuracy of eight previously proposed methods using prehistoric known-sex mummies from Northern Chile. J Forensic Sci 48(5):927–935PubMedCrossRef

31.

Vlak D, Roksandic M, Schillaci M (2008) Greater sciatic notch as a sex indicator in juveniles. Am J Phys Anthropol 137:309–315PubMedCrossRef

32.

Irurita J, Alemán I (2016) Validation of the sex estimation method elaborated by Schutkowski in the Granada Osteological Collection of identified infant and young children: analysis of the controversy between the different ways of analyzing and interpreting the results. Int J Legal Med 130:1623–1632CrossRef

33.

Mesejo P, Martos R, Ibáñez O, Novo J, Ortega M (2020) A survey on artificial intelligence techniques for biomedical image analysis in skeleton-based forensic human identification. Appl Sci 10:4703CrossRef

34.

Coelho J, Curate F (2019) CADOES: an interactive machine-learning approach for sex estimation with the pelvis. Forensic Sci Int 302:109873CrossRef

35.

Nikita E, Nikitas P (2020) On the use of machine learning algorithms in forensic anthropology. Legal Med 47:101771PubMedCrossRef

36.

Imaizumi K, Bermejo E, Taniguchi K, Ogawa Y, Nagata T, Kaga K, Hayakawa H, Shiotani S (2020) Development of a sex estimation method for skulls using machine learning on three-dimensional shapes of skulls and skull parts. Forensic Imaging 22:200393CrossRef

37.

Alemán I, Irurita J, A. valencia, A. Martínez, S. Lázaro, J Viciano, M Botella, (2012) Brief communication: the Granada Osteological Collection of identified infants and young children. Am J Phys Anthropol 149:606–610PubMedCrossRef

38.

Estévez E, López-Lázaro S, López-Morago C, Alemán I, Botella M (2017) Sex estimation of infants through geometric morphometric analysis of the ilium. Int J Legal Med 131:1747–1756PubMedCrossRef

39.

Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. IEEE Comput Soc Conf Comput Vis Pattern Recognit 1:886–893

40.

Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition, ICLR, p. arXiv: 1409.1556

41.

He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition, In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp 770–778

42.

Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20(3):273–297CrossRef

43.

Breiman L (2001) Randam forests. Mach Learn 45(1):5–32CrossRef

44.

Freund Y, Schapire R, Abe N (1999) A short introduction to boosting, Journal-Japanese Society For. Artif Intell 14(771–780):1612

45.

Deng J, Dong W, Socher R, Li L, Li K, Fei-Fei L (2009) Imagenet: a large-scale hierarchical image database. 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp 248–255. https://doi.org/10.1109/CVPR.2009.5206848

46.

Chollet F (2017) Deep learning with Python, Manning Publications

47.

Sutter R (2003) Nonmetric subadult skeletal sexing traits: I. A Blind test of the accuracy of eight previously proposed methods using prehistoric known-sex mummies from Northern Chile. J Forensic Sci 48(5):1–9CrossRef

48.

García-Mancuso R, Petrone S, Salceda S, González P (2018) Revisión crítica de la utilización del ilion para el diagnóstico de sexo en restos esqueléticos de individuos subadultos mediante técnicas morfométricas. Anales de Antropología 52(2):7–22CrossRef

49.

Marino R, Tanganelli V, Pietrobelli A, Belcastro MG (2021) Evaluation of the auricular surface method for subadult sex estimation on Italian modern (19th to 20th century) identified skeletal collections. Am J Phys Anthropol 174(4):792–803PubMedCrossRef

50.

Gama I, Navega D, Cunha E (2015) Sex estimation using the second cervical vertebra: a morphometric analysis in a documented Portuguese skeletal sample. Int J Legal Med 129:365–372PubMedCrossRef

51.

Curate F, Coelho J, Gonçalves D, Coelho C, Ferreira M, Navega D, Cunha E (2016) A method for sex estimation using the proximal femur. Forensic Sci Int 266:579.e1-579.e7CrossRef

52.

Ubelaker D (2008) Issues in the global applications of methodology in forensic anthropology. J Forensic Sci 53(3):606–607PubMedCrossRef

53.

Liversidge H (2015) Controversies in age estimation from developing teeth. Ann Hum Biol 42(4):397–406PubMedCrossRef

54.

Valsecchi A, Irurita J, Mesejo P (2019) Age estimation in forensic anthropology: methodological considerations about the validation studies of prediction models. Int J Legal Med 133:1915–1924PubMedCrossRef

55.

Buckberry J (2015) The (mis)use of adult age estimates in osteology. Ann Hum Biol 42(4):323–331PubMedCrossRef

56.

Lecun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278–2324CrossRef

57.

Lowe D (1999) Object recognition from local scale-invariant features. Proc Seventh IEEE Int Conf Comput Vis 2:1150–1157CrossRef

58.

Bay H, Ess A, Tuytelaars T, Van Gool L (2008) Speeded-up robust features (SURF). Compr Vision Image Underst 110(3):346–359CrossRef

59.

Castelvecchi D (2016) Can we open the black box of AI? Nature News 538(7623):20–23CrossRef

Title: Analysis of the performance of machine learning and deep learning methods for sex estimation of infant individuals from the analysis of 2D images of the ilium
Authors: Raúl Fernández Ortega
Javier Irurita
Enrique José Estévez Campo
Pablo Mesejo
Publication date: 01-11-2021
Publisher: Springer Berlin Heidelberg
Keyword: Artificial Intelligence
Published in: International Journal of Legal Medicine / Issue 6/2021
Print ISSN: 0937-9827
Electronic ISSN: 1437-1596
DOI: https://doi.org/10.1007/s00414-021-02660-6

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Analysis of the performance of machine learning and deep learning methods for sex estimation of infant individuals from the analysis of 2D images of the ilium

Abstract

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Abstract

Please log in to get access to this content

Other articles of this Issue 6/2021

DNA methylation of decedent blood samples to estimate the chronological age of human remains

Correction to: A distant relationship?—investigation of correlations between DNA isolated from backspatter traces recovered from firearms, wound profile characteristics, and shooting distance

Intra-bone nuclear DNA variability and STR typing success in Second World War first ribs

Correction to: Sharp force trauma with two katana swords: identifying the murder weapon by comparing tool marks on the skull bone

Practical application of synthetic head models in real ballistic cases

It is all about the insects: a retrospective on 20 years of forensic entomology highlights the importance of insects in legal investigations