Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/138316
Title: Detecció i diagnòstic de l'Alzheimer en imatges de ressonància magnètica
Author: Caravaca Müller, Oriol
Tutor: Nuñez Do Rio, Joan Manuel
Others: Ventura, Carles  
Abstract: Alzheimer's is the most common neurodegenerative disease worldwide and the most common cause of the development of dementia, where the detection of prodromal stages of Alzheimer's can help doctors around the world to prevent and improve the quality. of patients' lives. On the other hand, Neural Networks have been shown to be an effective means for the detection of disease disease MRI images. With this premise, we developed an ensamble of convolutional networks for the detection cases of Alzheimer's (AD) and its prodromal state called mild cognitive decline (MCI). Given the dimensionality of MRI volumes, each volume is subdivided into 45 regions of interest (ROI), and then top ten best performing regions are selected. Finally, for each of these 10 selected regions a classifier is trained and assembled to form the final model. With this methodology, the models obtained offer 80.9% accuracy in the Alzheimer's classifying task and 82% accuracy MCI predicting task.
Keywords: alzheimer
diagnosis
detection
Document type: info:eu-repo/semantics/bachelorThesis
Issue Date: 9-Jan-2022
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
ocaravacamTFG0122memòria.pdfMemòria del TFG1,1 MBAdobe PDFThumbnail
View/Open