Modelagem da dinâmica de um grupo de indivíduos HIV positivos com parâmetro Fuzzy do Tipo 2

• Ano de defesa: 2014
• Autor(a) principal: Castillo, Eduard Rojas
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Matemática; Ciências Exatas e da Terra; UFU
• Palavras-chave em Português: Conjuntos Fuzzy do tipo 2; Algoritmos Karnik-Mendel; Mancha de incerteza; HIV; Análise numérica; Teoria dos conjuntos; HIV (vírus); Type-2 Fuzzy sets; Algorithms Karnik-Mendel; Footprint of uncertainty; CNPQ::CIENCIAS EXATAS E DA TERRA::MATEMATICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/16810; https://doi.org/10.14393/ufu.di.2014.273

In this work we study the rate of return of a group of positive HIV (Human Immunodeficiency Virus) individuals, from symptomatic, if a person has opportunistic diseases, to asymptomatic, otherwise with antiretroviral treatment. We built a Fuzzy Rule-Based Systems (FRBS) for fuzzy sets of type-1 and type-2, in order to model the rate of return on positive HIV individuals of symptomatic to asymptomatic, depending on the viral load and CD4+ T lymphocyte level. The main lymphocyte attacked by the virus when it reaches the bloodstream is the CD4+ T lymphocyte. One of FRBS type-2 outputs is a real number and the other FRBS type-2 uses the Mamdani inference method and it has as output a singleton fuzzy set. From laboratory data CD4+ T lymphocyte and viral load in a group of positive HIV individuals, we obtain the values of the rates of return obtained by FRBS type-1 and the two FRBS of type-2, which are set by surfaces through the squares least method. The three surfaces are compared and we found that the two surfaces that best fit the values of the rates of return from symptomatic to asymptomatic are obtained by FRBS of type-2.