<Originals>Estimation of age from human renal tissues by image processing

概要

論文の詳細を見る
The purpose of this study was to estimate the age of unidentified dead subjects by examining renal tissue samples using image processing techniques and comparing the findings with standards obtained from subjects of known ages. Using human renal tissues, differences in the glomerulus and Bowman's capsule between ages were examined by image processing using a computer. Renal tissue samples collected from 400 male and female humans aged from less than 1 year to more than 60 years were treated by processed for microscopy. The sections were stained with hematoxylin and eosin. Since the changes with aging in the areas of the glomeruli and Bowman's capsules used for age estimation were linear, an estimation equation of age was determined by regression analysis. Estimation equation of age based on the area of glomeruli. Proportional coefficient, r=0.774 Y=0.319X-13.247. Estimation equation of age based on the area of Bowman's capsules. Proportional coefficient, r=0.830 Y=0.3293X-15.145. The accuracy of the estimated ages based on the area of Bowman's capsules were about 60% with a range of ±2 years, about 80% with a range of ±5 years, and about 95% with a range of ±8 years. The accuracy of the estimated ages based on the area of glomerulus were about 50% with a range of ±2 years, about 80% with a range of ±5 years, and about 95% with a range of ±8 years. These findings indicated to be useful for medicolegal studies.
近畿大学の論文