Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/133757
Title: Anàlisi de clustering per a l'exploració de dades biològiques multivariants
Author: Tomás Gascó, Anna
Director: Maceira, Marc  
Tutor: Fernández Martínez, Daniel  
Abstract: Cardiovascular disease is the leading cause of death worldwide. Among them, heart failure is a very common chronic pathology that is defined as the inability of the heart to function appropriately. In order to provide new knowledge in the field, it has been proposed to conduct a clustering study using a public database that collects clinical variables from 299 Pakistani patients over 40 years of age with heart failure. Five different methods of clustering analysis have been applied: k-means, agglomerative hierarchical clustering, hk-means, Gaussian mixture models and PAM with Gower distance. The first 4 have been applied using only the scaled numeric variables, but the last one has allowed us to use the whole dataset. In addition, a stratified hk-means study with the variables sex, survival, and ejection fraction has been included. The calculations were made using a value of k = 2, the most optimal number of clusters and the one that gave a more consistent result. With the baseline characteristics table for each cluster generated and the scatter plots, a consistent pattern could be found in which the clusters with a prevalence of patients who died during the study are characterized by having an average age and serum creatinine value higher than the cluster with patients who have survived. In short, this study provides a new perspective on data that has never been studied in this way. In addition, the conclusions reached with the classical methods of regression and survival have been reaffirmed.
Keywords: clustering
heart failure
multivariant data
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 16-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 
atomasgaTFM0621memòria.pdfMemòria del TFM1,66 MBAdobe PDFView/Open