Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/99167
Title: Detección y segmentación automática de lesiones en pacientes con esclerosis múltiple en imágenes de resonancia magnética
Author: García Pérez, José Carlos
Director: Casas-Roma, Jordi  
Tutor: Martínez de las Heras, Eloy
Abstract: Multiple sclerosis is a chronic neurodegenerative disease which affects the central nervous system. Magnetic resonance images allow the visualization and detection of brain lesions in patients with this disease, therefore monitoring such lesions is very important for both the diagnosis and monitoring of the disease, and to evaluate the effect of the treatment. Convolutional neural networks are currently a very promising technique to identify and automatically segment these lesions. This work aims to design and implement a convolutional neural network architecture to efficiently detect and segment brain lesions in magnetic resonance images, fine-tuning the hyperparameters or modifying the proposed pipeline in other existing architectures. The result of the process will be evaluated according to the Dice similarity coefficient on private data provided by the August Pi i Sunyer Biomedical Research Institute (IDIBAPS), which will allow to assess the accuracy and reproducibility of the model.
Keywords: convolutional neural networks
deep learning
brain lesions
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 9-Jun-2019
Publication license: http://creativecommons.org/licenses/by-nc-sa/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
jcarlosgarciaTFM0619memoria.pdfMemoria del TFM1,13 MBAdobe PDFThumbnail
View/Open
jcarlosgarciaTFM0619presentación.pdfPresentación del TFM754,4 kBAdobe PDFThumbnail
View/Open