Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Universidade de Lisb...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
addClaim

Estimação da creatinina sérica basal através de modelos GAMLSS

Authors: Mendonça, Inês Rodrigues;

Estimação da creatinina sérica basal através de modelos GAMLSS

Abstract

A doença renal aguda é caracterizada pela rápida diminuição da função renal. Esta patologia, por vezes, pode ser assintomática em alguns doentes, apresentando apenas variações de parâmetros laboratoriais que avaliam a taxa de filtração glomerular. O diagnóstico da lesão renal aguda é feito através de biomarcadores renais como a creatinina sérica. Este valor, juntamente com critérios de classificação da lesão renal aguda, consegue determinar a severidade da doença. No entanto, a classificação baseia-se em alterações da creatinina sérica em relação ao valor da creatinina sérica basal que frequentemente não está disponível. Assim, a abordagem utilizada na área clínica passa pela estimação dos valores da creatinina sérica basal através de varias fórmulas para o efeito. Neste estudo serão abordados três métodos de regressão, na tentativa de encontrar um modelo com bom poder preditivo que estime o valor da creatinina sérica basal. Assim sendo, o objetivo deste estudo passa pela aplicação de modelos lineares generalizados, modelos aditivos generalizados e modelos GAMLSS. Os modelos são avaliados pela sua capacidade preditiva através da analise dos resíduos. Uma vez que, a distribuição dos resíduos obtidos pelos modelos lineares generalizados e pelos modelos aditivos generalizados não foi normal, houve necessidade de prosseguir para uma nova abordagem baseada nos GAMLSS. Desta forma, foi alcançado o pressuposto de normalidade dos resíduos obtidos por estes modelos, embora as estimativas obtidas para os valores da creatinina sérica basal não tenham sido as melhores.

Acute renal disease is characterized by a rapid decrease of the renal function. This condition can, for some patients, be asymptomatic only changing for some laboratory parameters such as the glomerular filtration rate. The diagnosis of acute kidney injury is made by renal biomarkers such as serum creatinine. This value, along with the acute kidney injury classification criteria, can define the severity of the disease. However, the classification is based on changes in serum creatinine, when compared with a baseline value, which, most of the times, is not available. These values are replaced by estimates obtained through appropriate formulae. This study will present three regression approaches in an attempt to find a model with a good predictive power that is able to estimate the value of the baseline serum creatinine. Therefore, the goal of this work involves the application of generalized linear models, generalized additive models and GAMLSS models. These models are evaluated by their predictive ability, through a residuals analysis. The residuals of the generalized linear models and generalized additive models violated the assumption of normality and further analysis was needed, using the GAMLSS, to obtain a predictive model for the estimation of the baseline serum creatinine. With these models the normality of the residuals was achieved although the obtained estimates for baseline serum creatinine were not the best.

Tese de mestrado, Bioestatística, Universidade de Lisboa, Faculdade de Ciências, 2016

Country
Portugal
Related Organizations
Keywords

Teses de mestrado - 2016, Doença renal aguda, GAMLSS resíduos, Modelos lineares generalizados, Creatina sérica basal, Departamento de Estatística e Investigação Operacional, Modelos aditivos generalizados

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green