Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/133207
Title: Use of machine learning algorithms for analysing viral cure after antiretroviral treatment in HIV+ patients
Author: Pozo Rodríguez, Jordi del
Tutor: Perez-Alvarez, Nuria  
Others: Maceira, Marc  
Abstract: The main aim of this project was to apply survival analysis and machine learning algorithms to study viral cure in patients infected with Human immunodeficiency virus (HIV) from a clinical trial study after treatment with antiretroviral agents. One of the main challenges in this context is the presence of instances whose event results become unobservable after a certain moment, either due to an insufficiently long follow-up or because they did not present the event studied (called censorship). Currently, several machine learning algorithms adapted to analyse censored data are being developed. The objective of this Master thesis was to study three existing machine learning methods in the context described: Naïve Bayes, Artificial Neural Networks and Logistic regression and compare them with classical statistical methods. Towards this aim a database of a clinical trial has been used, containing real data on time to failure of antiretroviral treatment in patients infected with HIV+. The database required handling the missing data which was carried out by MICE algorithm. Two tested antiretroviral agents appear to have a similar effectiveness in treating HIV infection. After applying the selected machine learning algorithms to study viral cure, their performance was not higher than classfcal statistical models (Cox model), even after optimization. Nevertheless, the performance obtained with the three tested machine learning methods was high enough to consider further optimization of these algorithms in this field.
Keywords: survival analysis
machine learning
HIV
clinical trials
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 8-Jun-2021
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
joporoTFM0621memoria.pdfTFM Memory1,33 MBAdobe PDFThumbnail
View/Open