• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.3.2006.tde-14122006-153805
Document
Author
Full name
Nilson Tazawa
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2006
Supervisor
Committee
Justo Filho, Joao Francisco (President)
Ballester, Gerson
Hernandez, Emilio Del Moral
Title in Portuguese
Modelagem da dor utilizando-se redes neurais artificiais.
Keywords in Portuguese
Dor
Modelagem matemática
Redes neurais
Abstract in Portuguese
Este trabalho apresenta os resultados obtidos na elaboração de dois modelos para o fenômeno da dor utilizando-se redes neurais artificiais, de forma a simular computacionalmente as prováveis respostas de um indivíduo na presença de dor. Os modelos são fundamentados na Teoria de Controle da Comporta de dor, onde são analisados os principais componentes envolvidos na percepção/inibição da dor, bem como o funcionamento dos mecanismos biológicos e cognitivos participantes do processo. A escolha do tipo de rede neural é feita a partir das observações realizadas, considerando-se também o número de fatores envolvidos e o comportamento esperado frente a cada conjunto de entrada. O método de treinamento das redes neurais baseia-se no algoritmo de retropropagação. O foco do processamento da rede é responder adequadamente, considerando-se a influência das entradas envolvidas, a eventos posteriores à ocorrência de uma lesão gerando um sinal de alerta a ser utilizado como uma resposta natural do organismo a este dano tecidual. O desempenho de cada modelo é avaliado comparando-se as saídas obtidas com aquelas esperadas para cada padrão de entrada.
Title in English
Modeling of pain using artificial neural networks.
Keywords in English
Mathematical modeling
Neural networks
Pain
Abstract in English
This work presents the results of two models that were developed to describe the phenomenon of pain using Artificial Neural Networks, with the final goal to simulate computationally possible answers of an individual in the presence of pain. The models are based on The Gate Control Theory, where the main components involved in the perception/inhibition of pain were analysed, as well as the operation of the biological and cognitive mechanisms involved in the process. The type of neural network was chosen based on accumulated knowledge, considering also the number of involved factors and the expected behaviour response to each set of patterns. The neural networks were trained based on the backpropagation algorithm. The major focus of the network processing was to answer adequately to the occurrence of an injury, considering the role of the involved inputs, and generating an alert signal to this tecidual damage, which could replace the natural reply of the organism. The performance of each model is evaluated comparing the outputs obtained with those expected for each pattern.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2006-12-22
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.