Publicação

Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography

Ver documento

Detalhes bibliográficos
Resumo:Pregnancy still poses health risks that are not attended to by current clinical practice motorization procedures. Electrohysterography (EHG) record signals are analyzed in the course of this thesis as a contribution and effort to evaluate their suitability for pregnancy monitoring. The presented work is a contributes with an unsupervised classification solution for uterine contractile segments to FCT’s Uterine Explorer (UEX) project, which explores analysis procedures for EHG records. In a first part, applied processing procedures are presented and a brief exploration of the best practices for these. The procedures include those to elevate the representation of uterine events relevant characteristics, ease further computation requirements, extraction of contractile segments and spectral estimation. More detail is put into the study of which characteristics should be chosen to represent uterine events in the classification process and feature selection methods. To such end, it is presented the application of a principal component analysis (PCA) to three sets: interpolated contractile events, contractions power spectral densities, and to a number of computed features that attempt evidencing time, spectral and non-linear characteristics usually used in EHG related studies. Subsequently, a wrapper model approach is presented as a mean to optimize the feature set through cyclically attempting the removal and re-addition of features based on clustering results. This approach takes advantage of the fact that one class is known beforehand to use its classification accuracy as the criteria that defines whether the modification made to the feature set was ominous. Furthermore, this work also includes the implementation of a visualization tool that allows inspecting the effect of each processing procedure, the uterine events detected by different methods and clusters they were associated to by the final iteration of the wrapper model.
Autores principais:Morais, João Manuel de Oliveira Valente
Assunto:Electrohysterogram (EHG) Electromyogram (EMG) Unsupervised Classification Machine Learning Signal Processing
Ano:2019
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade Nova de Lisboa
Idioma:inglês
Origem:Repositório Institucional da UNL
_version_ 1868415392477609984
author Morais, João Manuel de Oliveira Valente
author_facet Morais, João Manuel de Oliveira Valente
author_role author
contributor_name_str_mv Batista, Arnaldo
RUN
country_str PT
creators_json_txt [{\"Person.name\":\"Morais, João Manuel de Oliveira Valente\"}]
datacite.contributors.contributor.contributorName.fl_str_mv Batista, Arnaldo
RUN
datacite.creators.creator.creatorName.fl_str_mv Morais, João Manuel de Oliveira Valente
datacite.date.Accepted.fl_str_mv 2019-09-01T00:00:00Z
datacite.date.available.fl_str_mv 2019-10-16T10:08:17Z
datacite.date.embargoed.fl_str_mv 2019-10-16T10:08:17Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_abf2
datacite.subjects.subject.fl_str_mv Electrohysterogram (EHG)
Electromyogram (EMG)
Unsupervised Classification
Machine Learning
Signal Processing
datacite.titles.title.fl_str_mv Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
dc.contributor.none.fl_str_mv Batista, Arnaldo
RUN
dc.creator.none.fl_str_mv Morais, João Manuel de Oliveira Valente
dc.date.Accepted.fl_str_mv 2019-09-01T00:00:00Z
dc.date.available.fl_str_mv 2019-10-16T10:08:17Z
dc.date.embargoed.fl_str_mv 2019-10-16T10:08:17Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10362/84451
dc.language.none.fl_str_mv eng
dc.rights.none.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.subject.none.fl_str_mv Electrohysterogram (EHG)
Electromyogram (EMG)
Unsupervised Classification
Machine Learning
Signal Processing
dc.title.fl_str_mv Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_bdcc
description Pregnancy still poses health risks that are not attended to by current clinical practice motorization procedures. Electrohysterography (EHG) record signals are analyzed in the course of this thesis as a contribution and effort to evaluate their suitability for pregnancy monitoring. The presented work is a contributes with an unsupervised classification solution for uterine contractile segments to FCT’s Uterine Explorer (UEX) project, which explores analysis procedures for EHG records. In a first part, applied processing procedures are presented and a brief exploration of the best practices for these. The procedures include those to elevate the representation of uterine events relevant characteristics, ease further computation requirements, extraction of contractile segments and spectral estimation. More detail is put into the study of which characteristics should be chosen to represent uterine events in the classification process and feature selection methods. To such end, it is presented the application of a principal component analysis (PCA) to three sets: interpolated contractile events, contractions power spectral densities, and to a number of computed features that attempt evidencing time, spectral and non-linear characteristics usually used in EHG related studies. Subsequently, a wrapper model approach is presented as a mean to optimize the feature set through cyclically attempting the removal and re-addition of features based on clustering results. This approach takes advantage of the fact that one class is known beforehand to use its classification accuracy as the criteria that defines whether the modification made to the feature set was ominous. Furthermore, this work also includes the implementation of a visualization tool that allows inspecting the effect of each processing procedure, the uterine events detected by different methods and clusters they were associated to by the final iteration of the wrapper model.
dirty 0
eu_rights_str_mv openAccess
format masterThesis
fulltext.url.fl_str_mv https://run.unl.pt/bitstreams/524a2ab3-fb60-403f-a81f-cbf5369ac514/download
id run_51112eccb6f65656a8d7d4a9f82dbbf4
identifier.url.fl_str_mv http://hdl.handle.net/10362/84451
instacron_str unl
institution Universidade Nova de Lisboa
instname_str Universidade Nova de Lisboa
language eng
network_acronym_str run
network_name_str Repositório Institucional da UNL
oai_identifier_str oai:run.unl.pt:10362/84451
organization_str_mv urn:organizationAcronym:unl
person_str_mv Morais, João Manuel de Oliveira Valente
publishDate 2019
reponame_str Repositório Institucional da UNL
repository_id_str urn:repositoryAcronym:run
service_str_mv urn:repositoryAcronym:run
spelling engpt_PTPregnancy still poses health risks that are not attended to by current clinical practice motorization procedures. Electrohysterography (EHG) record signals are analyzed in the course of this thesis as a contribution and effort to evaluate their suitability for pregnancy monitoring. The presented work is a contributes with an unsupervised classification solution for uterine contractile segments to FCT’s Uterine Explorer (UEX) project, which explores analysis procedures for EHG records. In a first part, applied processing procedures are presented and a brief exploration of the best practices for these. The procedures include those to elevate the representation of uterine events relevant characteristics, ease further computation requirements, extraction of contractile segments and spectral estimation. More detail is put into the study of which characteristics should be chosen to represent uterine events in the classification process and feature selection methods. To such end, it is presented the application of a principal component analysis (PCA) to three sets: interpolated contractile events, contractions power spectral densities, and to a number of computed features that attempt evidencing time, spectral and non-linear characteristics usually used in EHG related studies. Subsequently, a wrapper model approach is presented as a mean to optimize the feature set through cyclically attempting the removal and re-addition of features based on clustering results. This approach takes advantage of the fact that one class is known beforehand to use its classification accuracy as the criteria that defines whether the modification made to the feature set was ominous. Furthermore, this work also includes the implementation of a visualization tool that allows inspecting the effect of each processing procedure, the uterine events detected by different methods and clusters they were associated to by the final iteration of the wrapper model.application/pdfpt_PTUnsupervised Classification of Uterine Contractions Recorded Using ElectrohysterographyMorais, João Manuel de Oliveira ValenteBatista, ArnaldoHostingInstitutionOrganizationalRUNe-mailmailto:run@unl.ptrun@unl.pt2019-10-16T10:08:17Z2019-0920192019-09-01T00:00:00ZHandlehttp://hdl.handle.net/10362/84451http://purl.org/coar/access_right/c_abf2open accessElectrohysterogram (EHG)Electromyogram (EMG)Unsupervised ClassificationMachine LearningSignal Processing3328049 bytesliteraturehttp://purl.org/coar/resource_type/c_bdccmaster thesishttp://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://run.unl.pt/bitstreams/524a2ab3-fb60-403f-a81f-cbf5369ac514/download
spellingShingle Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
Morais, João Manuel de Oliveira Valente
Electrohysterogram (EHG)
Electromyogram (EMG)
Unsupervised Classification
Machine Learning
Signal Processing
status SINGLETON
subject.fl_str_mv Electrohysterogram (EHG)
Electromyogram (EMG)
Unsupervised Classification
Machine Learning
Signal Processing
title Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
title_full Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
title_fullStr Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
title_full_unstemmed Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
title_short Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
title_sort Unsupervised Classification of Uterine Contractions Recorded Using Electrohysterography
topic Electrohysterogram (EHG)
Electromyogram (EMG)
Unsupervised Classification
Machine Learning
Signal Processing
topic_facet Electrohysterogram (EHG)
Electromyogram (EMG)
Unsupervised Classification
Machine Learning
Signal Processing
url http://hdl.handle.net/10362/84451
visible 1