Exploration of Parkinson’s disease recognition space by artificial neural networks

Rojo Vicente, Mario (2022). Exploration of Parkinson’s disease recognition space by artificial neural networks. Proyecto Fin de Carrera / Trabajo Fin de Grado, E.T.S.I. de Sistemas Informáticos (UPM), Madrid.

Description

Title: Exploration of Parkinson’s disease recognition space by artificial neural networks
Author/s:
  • Rojo Vicente, Mario
Contributor/s:
  • Díaz Pérez, Francisco
Item Type: Final Project
Degree: Grado en Ingeniería del Software
Date: June 2022
Subjects:
Freetext Keywords: Machine learning; Parkinson; Data sets
Faculty: E.T.S.I. de Sistemas Informáticos (UPM)
Department: Sistemas Informáticos
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img] PDF - Users in campus UPM only - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (637kB)

Abstract

This paper focuses on the objective a Machine Learning model to classify the advancement of Parkinson´s Disease on a patient given some audio recordings and general information, such as age and gender. For this we will utilize a dataset extracted from synapse to train the several proposed models and finally extract conclusions on the outcomes. This project focuses on the implementation of CNN models based of the VGG-16 architecture. The results ranges from a 30% accuracy to around a 60% on all three datasets (train, validation and test) depending on the model and its hyper parameters, as well as the processes applied on the data. In conclusion the project was lacking more instances of reliable data but shows the possibilities of implementing such models on bigger datasets with rather significant results. The final results of this paper allow us to define rules and procedures to be implemented in similar future projects.

More information

Item ID: 71151
DC Identifier: https://oa.upm.es/71151/
OAI Identifier: oai:oa.upm.es:71151
Deposited by: Biblioteca Universitaria Campus Sur
Deposited on: 12 Jul 2022 19:21
Last Modified: 12 Jul 2022 19:21
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM