Elemental analysis of burnt human bone for classifying sex and age at death by logistic regression

Abstract

This study aims to develop a protocol for classifying sex and age groups of deceased persons by using the chemical information in burnt bone fragments. Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES) has been employed to assess the chemical compositions of the burnt bones from the deceased, all of Thai nationality. By performing three ICP-OES measurements on burnt bone samples from 75 individuals, 43 males and 32 females, with an accurate record of their age at death, only six elements i.e. Al, Ca, K, Mg, P, S, were observed in the fragments above the instrumental detection limit of ICP-OES. Fifteen concentration ratios of these particular elements were examined for their significant difference with respect to sex and age group by Mann–Whitney U and Kruskal–Wallis tests, respectively. Subsequently, the significantly different ratios were employed to assess the relationship between bone chemical compositions and sex/age groups of the deceased using logistic regression analysis. Binary and multinomial logistic regressions were required to classify sex and age group at death, respectively. Regression analyses indicated that the input variables representing sex should be included for classifying the age group of the deceased. The chemical information in burnt bone fragments could be employed to assess the sex and age groups of the deceased with 79.60% and 75.10% accuracy, respectively. Chemical measurements were performed on burnt bone samples from two additional individuals to verify the developed logistic equations and classification protocol. More fractions of bones from an individual give a higher percentage of correction for classifying sex and age at death of the deceased.