Please use this identifier to cite or link to this item:
Title: Implementation of an evaluation platform for Alzheimer patients based on Egocentric Sequences Description
Author: Soler Solé, Sergi
Director/Tutor: Bolaños Solà, Marc
Radeva, Petia
Keywords: Malaltia d'Alzheimer
Trastorns de la memòria
Treballs de fi de grau
Desenvolupament de programari d'aplicació
Xarxes neuronals (Informàtica)
Alzheimer's disease
Memory disorders
Computer software
Bachelor's thesis
Application software development
Neural networks (Computer science)
Issue Date: 30-Jan-2017
Abstract: Numerous international population-based studies have been conducted to document the frequency of MCI, estimating its prevalence to be between 15% and 20% in persons 60 years and older, making it a common condition encountered by clinicians[17]. This number is predicted to increase to 75.6 million in 2030, and 135.5 million in 2050[14], leading to deep social and economical costs. The most common dementia type is Alzheimer (between 50% and 70% of the cases) and its early detection can greatly affect the recovery of the patient. That is why it is important to have tools for its early diagnosis and follow-up. Serious games, with an increasing popularity, are a good way to MCI as an early stage of Alzheimer and improve the memory capacities of the patients. These video games focusing on different stages of the illness can help doctors to document and check the progress of the illness. This work aims on developing a software for patients with MCI, which is the lack of memory and other human characteristics like reasoning and language. These individuals usually progress to Alzheimer disease, but if detected early, in some cases they can also remain stable or even recover with time. To help them to exercise their memory, we propose that our program uses their own experiences caught by a wearable camera. This software will provide images of the patient’s life in order to do exercises that will evaluate their ability to remember and reason about the scenes they visualize. With these tests, the doctors will be able to see the evolution of the patients, and help them to diagnose and track the illness. In this project, we additionally work on an application, for the first time, of Deep Neural Networks for the automatic generation of descriptions of egocentric sequences. This will serve as the first step to automate the evaluation process by automatically comparing the subjective descriptions provided by the patients to the objective ones generated by our system.
Note: Treballs Finals de Grau d'Enginyeria Informàtica, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2017, Director: Marc Bolaños Solà i Petia Radeva
Appears in Collections:Programari - Treballs de l'alumnat
Treballs Finals de Grau (TFG) - Enginyeria Informàtica

Files in This Item:
File Description SizeFormat 
codi_font.zipCodi font504.45 MBzipView/Open
memoria.pdfMemòria33 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons