(5.86 MB) Abstract
Advisor: Verónica Vilaplana
Studies: Telecommunication Engineering at Telecom BCN-ETSETB from the Technical University of Catalonia (UPC)
Alzheimer’s disease currently affects more than 36 million people in the world. A patient’s brain suffers changes during the earliest stages of the disease and long before showing any clinical symptoms. For that reason, researchers focus their efforts towards defining which changes occur and where do they take place, with the goal of detecting indicators to predict the development of the disease. Specifically, the entity Fundación Pascual Maragall para la investigación contra el Alzheimer studies the processes of the brain all along the disease’s stages using images obtained through different MRI techniques. The huge volume of data generated in this kind of investigation is a big obstacle to carry out analysis and extracting conclusions. The aim of this thesis is making this process easier by using data mining techniques. The goal is to develop a basic classification system to distinguish in which stage of the disease a patient is in, using data extracted from cerebral images. This system must form the basis for a future data mining system that satisfies the necessities of the Fundación Pascual Maragall researchers. In addition to the classification system, this project focuses on distinguishing which is the most relevant data in the classification and on optimizing the classification in the pre-clinical stage of the disease.
