Abstract

ROSPIDE, M.A.  and  GOROSO, D.G.. Validation of an algorithm for calculation of tooth dimensions based on the method of Bloise. Cuad. med. forense [online]. 2014, vol.20, n.4, pp.180-190. ISSN 1988-611X.  https://dx.doi.org/10.4321/S1135-76062014000300004.

Introduction: When you study bone material from archaeological excavations or isolated remains, it is frequent that storage conditions are poor, so the anthropometric reconstruction of the main characteristic features cannot be observed. This problem is even greater in forensic cases in which, on many occasions, the identification must only be performed through tooth parts. For this reason, it is necessary to use validated alternative methods that allow the identification of an individual from dental fragmented remains. Goal: Validate the algorithm, based on the Bloise's method, by comparing the characteristic dimensions of a tooth in mesiodistal diameter known, taken as initial data for the calculation, with the calculated dimensions of the same tooth. Materials and methods: Software: an algorithm based on the Bloise's method in friendly Excel Version 14.0. 2010 was developed to calculate the size of the teeth from a single known piece. Instruments: the measurements were performed using a digital caliper, class 2, with 0.01 mm accuracy and a tooth gauge, class 2, with 0.01 mm accuracy. Subjects: were carried out 106 measures teeth, skulls and plaster models, were selected from Chair of Legal and Forensic Dentistry. Experimental procedure: a known tooth medial diameter was measured and its dimension was calculated using the algorithm. Analysis of data: the experimental data were compared with the values calculated by the algorithm using the method of Bland-Altman. For greater accuracy of the algorithm a correction factor was determined from the linear regression of Pearson coefficients (p <0.05). Finally, the corrected values and the experimental values were again compared using the method of Bland-Altman. Results: When the Bland-Altman's method was applied between the experimental and the calculated values without correction, it was observed a dispersion of up to two standard deviations. This problem was corrected using the correction factor determined by the Pearson coefficient, i.e. Cor = (Cal - 0.0183)/0.9601. This fix allowed to decrease the bias for differences of ±0.08 mm (less than one standard deviation) between the experimental value and the value obtained from the algorithm. Conclusion: Correction factors suggested to Bloise's method is closer to the dental data measured experimentally. From this correction, it was possible to validate the virtual simulator based on the Bloise's method.

Keywords : Tooth diameter; Bland-Altman; Validation of models.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )