A novel semi-supervised methodology for extracting tumor type-specific MRS sources in human brain data.

<h4>Background</h4>The clinical investigation of human brain tumors often starts with a non-invasive imaging study, providing information about the tumor extent and location, but little insight into the biochemistry of the analyzed tissue. Magnetic Resonance Spectroscopy can complement i...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Sandra Ortega-Martorell, Héctor Ruiz, Alfredo Vellido, Iván Olier, Enrique Romero, Margarida Julià-Sapé, José D Martín, Ian H Jarman, Carles Arús, Paulo J G Lisboa
Formato: article
Lenguaje:EN
Publicado: Public Library of Science (PLoS) 2013
Materias:
R
Q
Acceso en línea:https://doaj.org/article/cad2e552d41c4d049216a0d24458c15f
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:cad2e552d41c4d049216a0d24458c15f
record_format dspace
spelling oai:doaj.org-article:cad2e552d41c4d049216a0d24458c15f2021-11-18T08:40:34ZA novel semi-supervised methodology for extracting tumor type-specific MRS sources in human brain data.1932-620310.1371/journal.pone.0083773https://doaj.org/article/cad2e552d41c4d049216a0d24458c15f2013-01-01T00:00:00Zhttps://www.ncbi.nlm.nih.gov/pmc/articles/pmid/24376744/pdf/?tool=EBIhttps://doaj.org/toc/1932-6203<h4>Background</h4>The clinical investigation of human brain tumors often starts with a non-invasive imaging study, providing information about the tumor extent and location, but little insight into the biochemistry of the analyzed tissue. Magnetic Resonance Spectroscopy can complement imaging by supplying a metabolic fingerprint of the tissue. This study analyzes single-voxel magnetic resonance spectra, which represent signal information in the frequency domain. Given that a single voxel may contain a heterogeneous mix of tissues, signal source identification is a relevant challenge for the problem of tumor type classification from the spectroscopic signal.<h4>Methodology/principal findings</h4>Non-negative matrix factorization techniques have recently shown their potential for the identification of meaningful sources from brain tissue spectroscopy data. In this study, we use a convex variant of these methods that is capable of handling negatively-valued data and generating sources that can be interpreted as tumor class prototypes. A novel approach to convex non-negative matrix factorization is proposed, in which prior knowledge about class information is utilized in model optimization. Class-specific information is integrated into this semi-supervised process by setting the metric of a latent variable space where the matrix factorization is carried out. The reported experimental study comprises 196 cases from different tumor types drawn from two international, multi-center databases. The results indicate that the proposed approach outperforms a purely unsupervised process by achieving near perfect correlation of the extracted sources with the mean spectra of the tumor types. It also improves tissue type classification.<h4>Conclusions/significance</h4>We show that source extraction by unsupervised matrix factorization benefits from the integration of the available class information, so operating in a semi-supervised learning manner, for discriminative source identification and brain tumor labeling from single-voxel spectroscopy data. We are confident that the proposed methodology has wider applicability for biomedical signal processing.Sandra Ortega-MartorellHéctor RuizAlfredo VellidoIván OlierEnrique RomeroMargarida Julià-SapéJosé D MartínIan H JarmanCarles ArúsPaulo J G LisboaPublic Library of Science (PLoS)articleMedicineRScienceQENPLoS ONE, Vol 8, Iss 12, p e83773 (2013)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Sandra Ortega-Martorell
Héctor Ruiz
Alfredo Vellido
Iván Olier
Enrique Romero
Margarida Julià-Sapé
José D Martín
Ian H Jarman
Carles Arús
Paulo J G Lisboa
A novel semi-supervised methodology for extracting tumor type-specific MRS sources in human brain data.
description <h4>Background</h4>The clinical investigation of human brain tumors often starts with a non-invasive imaging study, providing information about the tumor extent and location, but little insight into the biochemistry of the analyzed tissue. Magnetic Resonance Spectroscopy can complement imaging by supplying a metabolic fingerprint of the tissue. This study analyzes single-voxel magnetic resonance spectra, which represent signal information in the frequency domain. Given that a single voxel may contain a heterogeneous mix of tissues, signal source identification is a relevant challenge for the problem of tumor type classification from the spectroscopic signal.<h4>Methodology/principal findings</h4>Non-negative matrix factorization techniques have recently shown their potential for the identification of meaningful sources from brain tissue spectroscopy data. In this study, we use a convex variant of these methods that is capable of handling negatively-valued data and generating sources that can be interpreted as tumor class prototypes. A novel approach to convex non-negative matrix factorization is proposed, in which prior knowledge about class information is utilized in model optimization. Class-specific information is integrated into this semi-supervised process by setting the metric of a latent variable space where the matrix factorization is carried out. The reported experimental study comprises 196 cases from different tumor types drawn from two international, multi-center databases. The results indicate that the proposed approach outperforms a purely unsupervised process by achieving near perfect correlation of the extracted sources with the mean spectra of the tumor types. It also improves tissue type classification.<h4>Conclusions/significance</h4>We show that source extraction by unsupervised matrix factorization benefits from the integration of the available class information, so operating in a semi-supervised learning manner, for discriminative source identification and brain tumor labeling from single-voxel spectroscopy data. We are confident that the proposed methodology has wider applicability for biomedical signal processing.
format article
author Sandra Ortega-Martorell
Héctor Ruiz
Alfredo Vellido
Iván Olier
Enrique Romero
Margarida Julià-Sapé
José D Martín
Ian H Jarman
Carles Arús
Paulo J G Lisboa
author_facet Sandra Ortega-Martorell
Héctor Ruiz
Alfredo Vellido
Iván Olier
Enrique Romero
Margarida Julià-Sapé
José D Martín
Ian H Jarman
Carles Arús
Paulo J G Lisboa
author_sort Sandra Ortega-Martorell
title A novel semi-supervised methodology for extracting tumor type-specific MRS sources in human brain data.
title_short A novel semi-supervised methodology for extracting tumor type-specific MRS sources in human brain data.
title_full A novel semi-supervised methodology for extracting tumor type-specific MRS sources in human brain data.
title_fullStr A novel semi-supervised methodology for extracting tumor type-specific MRS sources in human brain data.
title_full_unstemmed A novel semi-supervised methodology for extracting tumor type-specific MRS sources in human brain data.
title_sort novel semi-supervised methodology for extracting tumor type-specific mrs sources in human brain data.
publisher Public Library of Science (PLoS)
publishDate 2013
url https://doaj.org/article/cad2e552d41c4d049216a0d24458c15f
work_keys_str_mv AT sandraortegamartorell anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT hectorruiz anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT alfredovellido anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT ivanolier anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT enriqueromero anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT margaridajuliasape anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT josedmartin anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT ianhjarman anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT carlesarus anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT paulojglisboa anovelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT sandraortegamartorell novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT hectorruiz novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT alfredovellido novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT ivanolier novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT enriqueromero novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT margaridajuliasape novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT josedmartin novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT ianhjarman novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT carlesarus novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
AT paulojglisboa novelsemisupervisedmethodologyforextractingtumortypespecificmrssourcesinhumanbraindata
_version_ 1718421470078566400