Histomorphometric method for age-at-death estimation using the anterior side of the femoral midshaft^☆

Age-at-death estimation on poorly preserved human skeletons is a challenging task. Histological methods in most cases can overcome this limitation and provide accurate age-at-death estimates. Yet, the standard estimation error remains at high levels. This paper presents a newly developed histomorphometric method for age-at-death estimation using femoral midshaft cross-sections from individuals of know age and sex (n = 18). For the age-at-death estimation three non-linear polynomial regression equations based on the densities of bone microstructures were generated: (a) secondary osteon population density (SD), (b) osteon fragment population density (FD), and (c) haversian system density (HD). The validation of the regression equations was performed on a test series of six modern individuals of know age-at-death and sex. The applicability of the method was further tested on individuals of modern (n = 8) and archaeological origin (n = 39) whose age-at-death was estimated through standard anthropological methods. The HD regression equation provided the highest accuracy with standard estimation error of only ±3 years. The method proved applicable on skeletal elements with different level of preservation and reliable as it successfully estimated the age-at-death of the test series. Limitations related to the asymptote phenomenon and to pathological conditions have been observed and discussed.