Overview
Journal | 6th Annual ESHE Meeting |
Authors | Modesto Mata, Mario; García, Cecilia; Martín Francés, Laura; Martínez de Pinillos, Marina; García-González, Rebeca; Quintino, Yuliet; Canals, Antoni; Lozano, Marina; Dean, Christopher; Martinón Torres, María; Bermúdez de Castro, José María |
Volume | PESHE 5 |
Issue | |
Year | 2016 |
DOI | 10.13140/RG.2.2.27603.99361 |
Abstract
In the last years different methodologies have been developed to reconstruct worn teeth [1,2]. In this study we propose a new 2D methodology to reconstruct the worn enamel of lower molars in modern humans. Our main goal is to reconstruct molars with a high level of accuracy when measuring relevant histological variables and to validate the methodology calculating the errors associated with the measurement. To test whether this methodology is useful when applied to fossil teeth, we selected two unworn lower molars of Homo antecessor and reconstructed them by using the new methodology. This new methodology is based on polynomial regression equations and its accuracy has been validated measuring the crown height of the protoconid. This variable has been measured on a defined microcomputed tomography plane [3]. To perform the validation process in order to know the associated error of the measurements we followed the steps described by [4]. We also measured and calculated their associated errors following methodologies previously described [1,2]. Our sample consisted of a total of 25 unworn lower molars that belong to different historical and archaeological modern human populations from the Iberian Peninsula. 21 molars were used to perform the polynomial regression and 4 were used in the validation process. This second group of teeth were artificially worn and then reconstructed by using previously described methodologies and the new methodology based on polynomial regressions. We also tested if this polynomial model generated from a modern human sample was applicable to H. antecessor. To do it, we used two unworn lower molars that belong to this species. These molars were digitally worn and reconstructed with the polynomial regression based on modern humans, and compared their real and estimated crown height values. By using the new methodology, the mean percentage of error estimated in reconstructed molars for the crown height in comparison with its own real value is below 5%. This error significantly improves the results of other methodologies, both in the interobserver error and in the accuracy of the measurements. The interobserver error is also significantly lower than those obtained by other methods. When applied to H. antecessor, we observed that the percentage of error between the real and estimated values in the two unworn molars is 0.5% and 2.9%. The new methodology proposed in this study can be confidently applied to the reconstruction of lower molars of modern humans, as it improves the accuracy of the measurements and reduce the interobserver error.This new methodology can also be easily exportable to the human fossil record. Furthermore, we successfully applied our methodology to H. antecessor lower molars. By this way, we can confidently estimate real crown heights in slightly worn molars and thus divide their distances in deciles and assess perikymata packaging distribution.