Publicação

Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis

Ver documento

Detalhes bibliográficos
Resumo:Longitudinal cohort studies to study disease progression generally combine temporal features produced under periodic assessments (clinical follow-up) with static features associated with single-time assessments, genetic, psychophysiological, and demographic profiles. Subspace clustering, including biclustering and triclustering stances, enables the discovery of local and discriminative patterns from such multidimensional cohort data. These patterns, highly interpretable, are relevant to identifying groups of patients with similar traits or progression patterns. Despite their potential, their use for improving predictive tasks in clinical domains remains unexplored. In this work, we propose to learn predictive models from static and temporal data using discriminative patterns, obtained via biclustering and triclustering, as features within a state-of-the-art classifier, thus enhancing model interpretation. triCluster is extended to find time-contiguous triclusters in temporal data (temporal patterns) and a biclustering algorithm to discover coherent patterns in static data. The transformed data space, composed of bicluster and tricluster features, capture local and cross-variable associations with discriminative power, yielding unique statistical properties of interest. As a case study, we applied our methodology to follow-up data from Portuguese patients with Amyotrophic Lateral Sclerosis (ALS) to predict the need for non-invasive ventilation (NIV) since the last appointment. The results showed that, in general, our methodology outperformed baseline results using the original features. Furthermore, the bicluster/tricluster-based patterns used by the classifier can be used by clinicians to understand the models by highlighting relevant prognostic patterns.
Autores principais:Soares, Diogo F.
Outros Autores:Henriques, Rui; Gromicho, Marta; Carvalho, Mamede; Madeira, Sara C.
Assunto:Amyotrophic Lateral Sclerosis Biclustering Disease progression patterns Prognostic Three-way data Triclustering
Ano:2022
País:Portugal
Tipo de documento:artigo
Tipo de acesso:acesso aberto
Instituição associada:Universidade de Lisboa
Idioma:inglês
Origem:Repositório da Universidade de Lisboa
_version_ 1866810386625331200
author Soares, Diogo F.
author2 Henriques, Rui
Gromicho, Marta
Carvalho, Mamede
Madeira, Sara C.
author2_role author
author
author
author
author_facet Soares, Diogo F.
Henriques, Rui
Gromicho, Marta
Carvalho, Mamede
Madeira, Sara C.
author_role author
contributor_name_str_mv Repositório Científico de Acesso Aberto da ULisboa
country_str PT
creators_json_txt [{\"Person.name\":\"Soares, Diogo F.\",\"Person.identifier.orcid\":\"0000-0003-3503-0755\"},{\"Person.name\":\"Henriques, Rui\",\"Person.identifier.orcid\":\"0000-0002-3993-0171\"},{\"Person.name\":\"Gromicho, Marta\",\"Person.identifier.orcid\":\"0000-0003-2111-4579\"},{\"Person.name\":\"Carvalho, Mamede\",\"Person.identifier.orcid\":\"0000-0001-7556-0158\"},{\"Person.name\":\"Madeira, Sara C.\",\"Person.identifier.orcid\":\"0000-0002-1459-8096\"}]
datacite.contributors.contributor.contributorName.fl_str_mv Repositório Científico de Acesso Aberto da ULisboa
datacite.creators.creator.creatorName.fl_str_mv Soares, Diogo F.
Henriques, Rui
Gromicho, Marta
Carvalho, Mamede
Madeira, Sara C.
datacite.date.Accepted.fl_str_mv 2022-01-01T00:00:00Z
datacite.date.available.fl_str_mv 2022-09-16T15:25:21Z
datacite.date.embargoed.fl_str_mv 2022-09-16T15:25:21Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_abf2
datacite.subjects.subject.fl_str_mv Amyotrophic Lateral Sclerosis
Biclustering
Disease progression patterns
Prognostic
Three-way data
Triclustering
datacite.titles.title.fl_str_mv Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
dc.contributor.none.fl_str_mv Repositório Científico de Acesso Aberto da ULisboa
dc.creator.none.fl_str_mv Soares, Diogo F.
Henriques, Rui
Gromicho, Marta
Carvalho, Mamede
Madeira, Sara C.
dc.date.Accepted.fl_str_mv 2022-01-01T00:00:00Z
dc.date.available.fl_str_mv 2022-09-16T15:25:21Z
dc.date.embargoed.fl_str_mv 2022-09-16T15:25:21Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10451/54489
dc.language.none.fl_str_mv eng
dc.publisher.none.fl_str_mv Elsevier
dc.rights.cclincense.fl_str_mv http://creativecommons.org/licenses/by-nc-nd/4.0/
dc.rights.none.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.subject.none.fl_str_mv Amyotrophic Lateral Sclerosis
Biclustering
Disease progression patterns
Prognostic
Three-way data
Triclustering
dc.title.fl_str_mv Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_6501
description Longitudinal cohort studies to study disease progression generally combine temporal features produced under periodic assessments (clinical follow-up) with static features associated with single-time assessments, genetic, psychophysiological, and demographic profiles. Subspace clustering, including biclustering and triclustering stances, enables the discovery of local and discriminative patterns from such multidimensional cohort data. These patterns, highly interpretable, are relevant to identifying groups of patients with similar traits or progression patterns. Despite their potential, their use for improving predictive tasks in clinical domains remains unexplored. In this work, we propose to learn predictive models from static and temporal data using discriminative patterns, obtained via biclustering and triclustering, as features within a state-of-the-art classifier, thus enhancing model interpretation. triCluster is extended to find time-contiguous triclusters in temporal data (temporal patterns) and a biclustering algorithm to discover coherent patterns in static data. The transformed data space, composed of bicluster and tricluster features, capture local and cross-variable associations with discriminative power, yielding unique statistical properties of interest. As a case study, we applied our methodology to follow-up data from Portuguese patients with Amyotrophic Lateral Sclerosis (ALS) to predict the need for non-invasive ventilation (NIV) since the last appointment. The results showed that, in general, our methodology outperformed baseline results using the original features. Furthermore, the bicluster/tricluster-based patterns used by the classifier can be used by clinicians to understand the models by highlighting relevant prognostic patterns.
dirty 0
eu_rights_str_mv openAccess
format article
fulltext.url.fl_str_mv https://repositorio.ulisboa.pt/bitstreams/4cf2cf6b-b2df-4f00-8cb6-988af3b90d11/download
funding.funder.alternateName_str_mv FCT
FCT
FCT
FCT
EC
funding.funder.identifier_str_mv http://doi.org/10.13039/501100001871
http://doi.org/10.13039/501100001871
http://doi.org/10.13039/501100001871
http://doi.org/10.13039/501100001871
http://doi.org/10.13039/501100008530
funding.funder.name_str_mv Fundação para a Ciência e a Tecnologia
Fundação para a Ciência e a Tecnologia
Fundação para a Ciência e a Tecnologia
Fundação para a Ciência e a Tecnologia
European Commission
funding.name_str_mv 6817 - DCRRNI ID
6817 - DCRRNI ID
6817 - DCRRNI ID
H2020
id ul_4cb362b8fded8d0ea8dcf8fec493db1f
identifier.url.fl_str_mv http://hdl.handle.net/10451/54489
instacron_str ul
institution Universidade de Lisboa
instname_str Universidade de Lisboa
language eng
network_acronym_str ul
network_name_str Repositório da Universidade de Lisboa
oai_identifier_str oai:repositorio.ulisboa.pt:10451/54489
organization_str_mv urn:organizationAcronym:ul
person_str_mv Soares, Diogo F.
Soares, Diogo F.
https://www.ciencia-id.pt/D713-C08C-A84A
D713-C08C-A84A
http://orcid.org/0000-0003-3503-0755
0000-0003-3503-0755
Henriques, Rui
Henriques, Rui
https://www.ciencia-id.pt/3D1F-D220-28D1
3D1F-D220-28D1
http://orcid.org/0000-0002-3993-0171
0000-0002-3993-0171
Gromicho, Marta
Gromicho, Marta
https://www.ciencia-id.pt/EA1D-1979-5C3A
EA1D-1979-5C3A
http://orcid.org/0000-0003-2111-4579
0000-0003-2111-4579
Carvalho, Mamede
Carvalho, Mamede
http://orcid.org/0000-0001-7556-0158
0000-0001-7556-0158
Madeira, Sara C.
Madeira, Sara C.
https://www.ciencia-id.pt/AF12-AA0D-0C7B
AF12-AA0D-0C7B
http://orcid.org/0000-0002-1459-8096
0000-0002-1459-8096
publishDate 2022
publisher.none.fl_str_mv Elsevier
reponame_str Repositório da Universidade de Lisboa
repository_id_str urn:repositoryAcronym:ul
service_str_mv urn:repositoryAcronym:ul
spelling engElsevierpt_PTLongitudinal cohort studies to study disease progression generally combine temporal features produced under periodic assessments (clinical follow-up) with static features associated with single-time assessments, genetic, psychophysiological, and demographic profiles. Subspace clustering, including biclustering and triclustering stances, enables the discovery of local and discriminative patterns from such multidimensional cohort data. These patterns, highly interpretable, are relevant to identifying groups of patients with similar traits or progression patterns. Despite their potential, their use for improving predictive tasks in clinical domains remains unexplored. In this work, we propose to learn predictive models from static and temporal data using discriminative patterns, obtained via biclustering and triclustering, as features within a state-of-the-art classifier, thus enhancing model interpretation. triCluster is extended to find time-contiguous triclusters in temporal data (temporal patterns) and a biclustering algorithm to discover coherent patterns in static data. The transformed data space, composed of bicluster and tricluster features, capture local and cross-variable associations with discriminative power, yielding unique statistical properties of interest. As a case study, we applied our methodology to follow-up data from Portuguese patients with Amyotrophic Lateral Sclerosis (ALS) to predict the need for non-invasive ventilation (NIV) since the last appointment. The results showed that, in general, our methodology outperformed baseline results using the original features. Furthermore, the bicluster/tricluster-based patterns used by the classifier can be used by clinicians to understand the models by highlighting relevant prognostic patterns.application/pdfpt_PTLearning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosisPersonalSoares, Diogo F.DSpacehttp://dspace.org/items/1b0f1562-0792-4cdc-b787-e624c71acd06DSpacehttp://dspace.org/items/1b0f1562-0792-4cdc-b787-e624c71acd06SoaresDiogoCiência IDhttps://www.ciencia-id.ptD713-C08C-A84AORCIDhttp://orcid.org0000-0003-3503-0755PersonalHenriques, RuiDSpacehttp://dspace.org/items/e9b43d60-75aa-48ca-88d5-45a85b8d8b7dDSpacehttp://dspace.org/items/e9b43d60-75aa-48ca-88d5-45a85b8d8b7dHenriquesRuiCiência IDhttps://www.ciencia-id.pt3D1F-D220-28D1ORCIDhttp://orcid.org0000-0002-3993-0171PersonalGromicho, MartaDSpacehttp://dspace.org/items/c6f0ac72-fdd9-499d-87e2-f13aa1545565DSpacehttp://dspace.org/items/c6f0ac72-fdd9-499d-87e2-f13aa1545565SilvaMarta Luísa Gromicho MorgadoCiência IDhttps://www.ciencia-id.ptEA1D-1979-5C3AORCIDhttp://orcid.org0000-0003-2111-4579Scopus Author IDhttps://www.scopus.com6506583274PersonalCarvalho, MamedeDSpacehttp://dspace.org/items/dd7f55d4-c2b5-4fd2-9bd1-a9542a62f58fDSpacehttp://dspace.org/items/dd7f55d4-c2b5-4fd2-9bd1-a9542a62f58fde CarvalhoMamedeORCIDhttp://orcid.org0000-0001-7556-0158Scopus Author IDhttps://www.scopus.com7101893769PersonalMadeira, Sara C.DSpacehttp://dspace.org/items/f6c1f830-a944-47e3-a471-51014c18932eDSpacehttp://dspace.org/items/f6c1f830-a944-47e3-a471-51014c18932eMadeiraSaraCiência IDhttps://www.ciencia-id.ptAF12-AA0D-0C7BORCIDhttp://orcid.org0000-0002-1459-8096Researcher IDhttps://www.researcherid.comC-5494-2008Scopus Author IDhttps://www.scopus.com6602138051HostingInstitutionOrganizationalRepositório Científico de Acesso Aberto da ULisboae-mailmailto:repositorio@reitoria.ulisboa.ptrepositorio@reitoria.ulisboa.ptISSNIsPartOf1532-0464DOIIsPartOf10.1016/j.jbi.2022.1041722022-09-16T15:25:21Z20222022-01-01T00:00:00ZHandlehttp://hdl.handle.net/10451/54489http://purl.org/coar/access_right/c_abf2open accessAmyotrophic Lateral SclerosisBiclusteringDisease progression patternsPrognosticThree-way dataTriclustering2195742 bytesFundação para a Ciência e a TecnologiaLASIGE - Extreme Computing6817 - DCRRNI IDCrossref Funder IDhttp://doi.org/10.13039/501100001871Fundação para a Ciência e a TecnologiaLASIGE - Extreme Computing6817 - DCRRNI IDCrossref Funder IDhttp://doi.org/10.13039/501100001871Fundação para a Ciência e a TecnologiaInstituto de Engenharia de Sistemas e Computadores, Investigação e Desenvolvimento em Lisboa6817 - DCRRNI IDCrossref Funder IDhttp://doi.org/10.13039/501100001871Fundação para a Ciência e a TecnologiaEfficient Algorithms for Temporal Triclustering: Targeting the Challenges of Temporality for Effective Solutions in Biomedical ApplicationsCrossref Funder IDhttp://doi.org/10.13039/501100001871European CommissionBRinging Artificial INTelligencE home for a better cAre of amyotrophic lateral sclerosis and multiple SclERosisH2020Crossref Funder IDhttp://doi.org/10.13039/501100008530literaturehttp://purl.org/coar/resource_type/c_6501journal article2022http://creativecommons.org/licenses/by-nc-nd/4.0/http://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://repositorio.ulisboa.pt/bitstreams/4cf2cf6b-b2df-4f00-8cb6-988af3b90d11/downloadJournal of Biomedical Informatics134
spellingShingle Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
Soares, Diogo F.
Amyotrophic Lateral Sclerosis
Biclustering
Disease progression patterns
Prognostic
Three-way data
Triclustering
status SINGLETON
subject.fl_str_mv Amyotrophic Lateral Sclerosis
Biclustering
Disease progression patterns
Prognostic
Three-way data
Triclustering
title Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
title_full Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
title_fullStr Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
title_full_unstemmed Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
title_short Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
title_sort Learning prognostic models using a mixture of biclustering and triclustering: predicting the need for non-invasive ventilation in amyotrophic lateral sclerosis
topic Amyotrophic Lateral Sclerosis
Biclustering
Disease progression patterns
Prognostic
Three-way data
Triclustering
topic_facet Amyotrophic Lateral Sclerosis
Biclustering
Disease progression patterns
Prognostic
Three-way data
Triclustering
url http://hdl.handle.net/10451/54489
visible 1