A Novel Method for Race Determination of Human Skulls
Race determination of skulls of individuals is a continually growing subject in forensic anthropology. Traditionally, race determination has been conducted either entirely subjectively by qualified forensic anthropologists, or has been conducted through a semi-automated fashion through multivariate discriminant functions. This paper describes a novel method for completely automated race determination of CT scans of skulls, wherein skulls are preprocessed, reduced to a low dimensional model and segregated into one of two racial classes through a classifier. The classifier itself is chosen from a survey conducted against four different classification techniques. This method can both be used as a tool for completely automated race determination, or as decision support for forensic anthropologists. A total of 341 skulls with variance in race have been gathered by the University of Nottingham Malaysia Campus and used to train and test the method. The resultant accuracy of this method is 79%.
The authors would like to thank the Hospital Kuala Lumpur for providing all data used. The research (NMRR-15-1761-2777) has received full ethics approval from the Ministry of Health, Malaysia.
- 3.National Institute of Health: Anthropological views. Accessed 06 Apr 2017Google Scholar
- 10.Jantz, R.L., Mahfouz, M., Shirley, N.R., Fatah, E.A.: Improving sex estimation from crania using 3-dimensional CT scans. Department of Justice (2013)Google Scholar
- 11.Besl, P.J., McKay, N.D.: Method for registration of 3-D shapes. In: Robotics-DL Tentative, pp. 586–606. International Society for Optics and Photonics (1992)Google Scholar
- 13.Rusinkiewicz, S., Levoy, M.: Efficient variants of the ICP algorithm. In: 2001 Proceedings of Third International Conference on 3-D Digital Imaging and Modeling, pp. 145–152. IEEE (2001)Google Scholar
- 14.Vos, F.M., et al.: A statistical shape model without using landmarks. In: 2004 Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004, vol. 3, pp. 714–717. IEEE (2004)Google Scholar
- 20.Fodor, I.K.: A survey of dimension reduction techniques. Cent. Appl. Sci. Comput. Lawrence Livermore Natl. Lab. 9, 1–18 (2002)Google Scholar
- 23.Fleute, M., Lavallée, S.: Building a complete surface model from sparse data using statistical shape models: application to computer assisted knee surgery. In: Wells, W.M., Colchester, A., Delp, S. (eds.) MICCAI 1998. LNCS, vol. 1496, pp. 879–887. Springer, Heidelberg (1998). https://doi.org/10.1007/BFb0056276CrossRefGoogle Scholar
- 25.Luo, L., et al.: Automatic sex determination of skulls based on a statistical shape model. Comput. Math. Methods Med. 2013 (2013)Google Scholar
- 28.Liaw, A., Wiener, M.: Classification and regression by randomforest. R news 2(3), 18–22 (2002)Google Scholar
- 29.Kobbelt, L., Campagna, S., Seidel, H.-P.: A general framework for mesh decimation. In: Graphics Interface, vol. 98, pp. 43–50 (1998)Google Scholar
- 34.Kohavi, R., et al.: A study of cross-validation and bootstrap for accuracy estimation and model selection. In: IJCAI, Stanford, CA, vol. 14, pp. 1137–1145 (1995)Google Scholar
- 35.Cranial features and race. http://johnhawks.net/explainer/laboratory/race-cranium/. Accessed 04 Apr 2017