الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 248 |  |  |
| | | from | 248 | | |
Abstract
In modern forensic situations, stature estimation can identify dead unknown persons. Skeletal stature is estimated in forensic anthropology as a component of a biological profile that also includes factors like ancestry, sex, and age.
The anatomical and mathematical approaches are the two primary techniques for stature estimation. Since the anatomical technique takes into account variations in long bone proportionality between groups, it is the most appropriate method for all population groups. In mathematical method, a prediction formula for stature prediction is often provided by long bone lengths regressed on stature.
In addition to being population-specific, most of stature formulae are also period-specific. Secular transition, a transient generational change in human morphology, affects stature. So, the aim of the present study was to analyze percutaneous and radiographic measurements of different human bones (hand, ulna, tibia, and foot bones) and their relationship to stature in order to make different regression formulae to estimate stature via the use of statistical analysis in a cluster random sample, in Menoufia governorate.
➢
from the current study the following results can be concluded:
Regarding sex, female participants represented (53.5%) and males represented (46.5%). The mean age of all studied participants was 32.49±6.30 years. Subjects from rural areas were more frequent than those from urban areas. The studied participants represented different occupations and different socioeconomic standard.
The mean stature of the studied participants was 166.24±9.15 cm, males were taller than females (the mean stature were 173.87±5.88 cm and
Summary & Conclusion
165
159.60±5.63 cm respectively). There were significant higher mean values for all percutaneous and radiological measures in males than females.
Regarding Pearson correlation, there were high significant positive correlations between stature with percutaneous measures in the studied participants of both upper limbs including hand length, palm width, and ulnar length and lower limb including tibial length, foot length and foot width, where the P value was <0.001 for all these measures, and there was significant positive correlation between stature with percutaneous measures of palm length of the studied participants, where the P value was <0.05.
Tibial length had the highest value of Pearson correlation coefficient, so it was the best parameter to predict stature either percutaneously or radiologically measured (r = 0.70, 0.76 respectively), followed by ulnar length.
In the present work, ulnar length was the best radiologically measured parameter of the studied bones in upper limb, and the second radiologically measured parameter after tibial length with strong correlation coefficient (r= 0.73).
As regard percutaneous measures, the second-best predictor was foot length with strong correlation coefficient (r = 0.69), followed by hand length (r = 0.65), palm width (r =0.57) and palm length (r =0.38).
In radiologically measured parameters in the present work, the second-best predictor of the studied parameters in upper limb was hand length (r = 0.64), followed by palm length and palm width (r = 0.55 for both).
In the current study, hand length -either measured percutaneously or radiologically- showed strong positive correlation with female stature, while in males showed moderate correlation.
Summary & Conclusion
166
Foot width was the least parameter to estimate stature either percutaneously or radiologically measured with a weak correlation coefficient (r = 0.20, 0.35 respectively).
from the result of the present work, different simple regression equations for stature estimation had been concluded using percutaneous and radiological parameters in the upper and lower limb with a highly significant P value. So, simple linear regression equation using single parameter could significantly be used for stature estimation. Simple regression equation using tibial length, foot length, ulnar length and hand length showed the least SEE. The SEE of the regression equation using tibial length either percutaneously or radiologically measured was 0.122 and 0.104 respectively.
from the present work, multiple linear regression formulae showed higher accuracy than simple linear regression equations, so multiple linear regression equations were better for stature estimation.
Multiple linear regression equation for stature prediction using all studied parameters (measured percutaneously) was S=54.168+1.863xHL-2.737x PL+2.537xPW+0.537xUL+0.920xTL+1.628xFL-0.853xFW, with accuracy of 70%.
In radiologically measured parameters, the multiple linear regression formula for stature estimation of overall subjects was recorded as S=58.533+1.716xHL-1.491xPL+2.244xPW+0.928xUL+1.445xTL-0.311xFL-0.569xFW. The accuracy of the equation was 66%.
By using parameters of upper and lower limb separately, multiple linear regression equation using percutaneous parameters in lower limb was better for stature estimation than that of upper limb parameters (R2 = 0.623 and 0.597 respectively).
Summary & Conclusion
167
The present results concluded that, there was no significant difference between measured stature and stature estimated by multiple linear regression equations for percutaneous or radiological measures in the studied participants (p>0.05). The mean measured stature was 166.24±9.15 cm, the percutaneous estimated stature was 166.04±7.63 cm, and the radiological estimated stature was 166.25±7.41 cm.
In the present work, there was no significant difference between measured stature and stature estimated by multiple linear regression equations for percutaneous or radiological measures in each of the upper and the lower limb parameters in the studied participants.
Regarding the result of the present study, tibial length / stature ratio was 23.05% ± 1.17 in all participants, 22.91% ± 1.13 in males, and in females was 23.17% ± 1.20.
Ulnar length /stature ratio for all studied participants was 17.14% ±1.00, while in males the mean percentages was17.17% ± 0.95 decreased to 17.11% ± 1.04 in females.