Bibliographic citations
Llajaruna, G., (2011). Automatización de la Datación de Equimosis en el Peritaje Médico Legal Peruano mediante Redes Neuronales Artificiales y Procesamiento de Imágenes [Tesis, Universidad Nacional Mayor de San Marcos]. https://hdl.handle.net/20.500.12672/1071
Llajaruna, G., Automatización de la Datación de Equimosis en el Peritaje Médico Legal Peruano mediante Redes Neuronales Artificiales y Procesamiento de Imágenes [Tesis]. PE: Universidad Nacional Mayor de San Marcos; 2011. https://hdl.handle.net/20.500.12672/1071
@misc{renati/488839,
title = "Automatización de la Datación de Equimosis en el Peritaje Médico Legal Peruano mediante Redes Neuronales Artificiales y Procesamiento de Imágenes",
author = "Llajaruna Pereda, Gisella Marilú",
publisher = "Universidad Nacional Mayor de San Marcos",
year = "2011"
}
The ecchymosis has a special medical - legal interest, as these provide the most valuable clues for the reconstruction of violence in which they occurred. Traditionally, doctors according to their experience using visual assessment to determine the age of ecchymosis, but this technique has been substantially subjective and has proven to be inaccurate and unreliable. The purpose of this thesis is to develop an intelligent system that incorporates image processing technique using the RGB colorimetric methodology in order to obtain the average pixel of indurated area of ecchymosis and a multilayer perceptron network whose input variables the average color of the indurated area obtained by images processing, color, age and sex of the injured, the presence of loose tissue, bone tissue and vascular tissue, as well as the ambient temperature, which have been carefully selected in order to obtain a more accurate dating of ecchymosis. The image processing is implemented in the Java programming language, which aims to obtain the average of the RGB pixels most influential of indurated area of ecchymosis. Also, the learning and validation phase of the multilayer perceptron network was performed with the mathematical tool MATLAB, using the backpropagation algorithm which provided an error rate of 1.26% and 1.37%, respectively. In consequence, the intelligent system proposed in this thesis generates a more precise dating ecchymosis compared to 80% of incorrect medical diagnoses. Keywords: Dating of ecchymosis, Image processing, Backpropagation.
This item is licensed under a Creative Commons License