Please use this identifier to cite or link to this item:
Title: Aplicación de métodos de aprendizaje automático para el estudio del análisis de supervivencia en pacientes infectados por el VIH
Author: Navarrete Bellot, Luis
Director: Ventura Royo, Carles  
Tutor: Pérez Álvarez, Núria
Keywords: hiv/aids
machine learning
survival analysis
Issue Date: Jun-2020
Publisher: Universitat Oberta de Catalunya (UOC)
Abstract: Survival analysis aims to analyze and model data where the result is the time until an event of interest occurs. One of the main challenges in this context is the presence of instances whose event results become unobservable after a certain moment, either because there is not a long enough follow-up or because they did not present the studied event (called censorship). Many adapted machine learning algorithms are currently being developed to analyze censored data. Thus, the objective of the TFM is defined to study the existing methods of machine learning in the described context, to find the most suitable method / s to study the time to failure of antiretroviral treatment in a cohort of patients infected with HIV. This project seeks to create description tools that allow summarizing the information contained in the data and generating a model that helps us classify patients according to the distribution of time until virological failure. To do this, we have the 'dsurv' database from the 'Instinct' study, which contains 342 variables from 995 patients who have been followed and evaluated at different points in time, giving rise to an unbalanced and heterogeneous base. In the project, we will analyze how the type of treatment that the patient receives and the performance of tests for mutations in integration, protection and reverse transcription can influence virological failure, or increase the probability of survival over time.
Language: Spanish
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
lnavarretebTFM0620memoria.pdfMemoria del TFM1.46 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons