Estimating Parkinson’s Disease Stages from Accelerometer Data

Detalhes bibliográficos
Autor(a) principal: Lobo, Vitor Marques
Data de Publicação: 2022
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10451/57677
Resumo: Tese de mestrado, Engenharia Informática (Engenharia de Software), 2022, Universidade de Lisboa, Faculdade de Ciências
id RCAP_3f5d3abff5e5408175b9718c22252501
oai_identifier_str oai:repositorio.ul.pt:10451/57677
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Estimating Parkinson’s Disease Stages from Accelerometer DataDoença de ParkinsonMarchaMachine LearningTeses de mestrado - 2022Departamento de InformáticaTese de mestrado, Engenharia Informática (Engenharia de Software), 2022, Universidade de Lisboa, Faculdade de CiênciasCurrent practices for monitoring disease stage and progression in Parkinson’s Disease still rely on periodic clinical visitations that can be cumbersome and highly stressful for patients and caretakers. Furthermore, the current gold standard for these evaluations is still reliant on observations made by trained clinicians using clinical scales that, in spite of their repeatedly verified validity, are subject to fluctuations due to intra or inter-rater variability. Over the last decade, technological developments in sensors for data collection and data science algorithms have enabled systems and tools for health and tele-medicine applications, along with a battery of research into the objective and continuous monitoring of Parkinson’s Disease. Among such research, gait and it’s characteristics have emerged as reliable markers for the progression of PD. As such, studies leveraging these characteristics for the objective monitoring of the disease have become a common trend in the related literature. A limiting factor in several of these studies is the use of scores and outcomes that, in spite of their high correlation to established clinical scales, are different to those usually used by clinicians, making them harder to interpret and adopt for clinical use. To bridge this gap, several studies have attempted to classify or estimate specific parts of the MDS-UPDRS, the most clinically used scale for the assessment of PD. Recently, the automatic estimation of scores for part 3 of the MDS-UPDRS, which focuses the severity of motor symptoms, have leveraged the use of deep learning techniques for this purpose, with promising results. This work presents a comparison of traditional feature engineered models against those current state of the art deep learning approaches for the prediction of this score. Furthermore, an analysis on different approaches to data collection, feature extraction and model parametrization was also performed, to assess the effect of these different variables on the estimation task. Finally, an analysis of the best configurations of machine learning pipelines for this purpose was also performed to direct further studies. While the optimal models in this study failed to match the performance of those state of the art approaches, the identified limitations and recommendations for future research form a solid foundation for future work on this topic.Guerreiro, TiagoRepositório da Universidade de LisboaLobo, Vitor Marques2023-05-30T10:50:03Z202220222022-01-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10451/57677enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-11-08T17:06:29Zoai:repositorio.ul.pt:10451/57677Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T22:08:13.469857Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv Estimating Parkinson’s Disease Stages from Accelerometer Data
title Estimating Parkinson’s Disease Stages from Accelerometer Data
spellingShingle Estimating Parkinson’s Disease Stages from Accelerometer Data
Lobo, Vitor Marques
Doença de Parkinson
Marcha
Machine Learning
Teses de mestrado - 2022
Departamento de Informática
title_short Estimating Parkinson’s Disease Stages from Accelerometer Data
title_full Estimating Parkinson’s Disease Stages from Accelerometer Data
title_fullStr Estimating Parkinson’s Disease Stages from Accelerometer Data
title_full_unstemmed Estimating Parkinson’s Disease Stages from Accelerometer Data
title_sort Estimating Parkinson’s Disease Stages from Accelerometer Data
author Lobo, Vitor Marques
author_facet Lobo, Vitor Marques
author_role author
dc.contributor.none.fl_str_mv Guerreiro, Tiago
Repositório da Universidade de Lisboa
dc.contributor.author.fl_str_mv Lobo, Vitor Marques
dc.subject.por.fl_str_mv Doença de Parkinson
Marcha
Machine Learning
Teses de mestrado - 2022
Departamento de Informática
topic Doença de Parkinson
Marcha
Machine Learning
Teses de mestrado - 2022
Departamento de Informática
description Tese de mestrado, Engenharia Informática (Engenharia de Software), 2022, Universidade de Lisboa, Faculdade de Ciências
publishDate 2022
dc.date.none.fl_str_mv 2022
2022
2022-01-01T00:00:00Z
2023-05-30T10:50:03Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10451/57677
url http://hdl.handle.net/10451/57677
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799134636402540544

Registros relacionados