Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 13
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Compatibility between 3T 1H SV-MRS data and automatic brain tumour diagnosis support systems based on databases of 1.5T 1H SV-MRS spectra

  • Fuster García, Elíes|||0000-0002-0716-8960
  • Garcia-Gomez, Juan M
  • Sáez Silvestre, Carlos|||0000-0003-2678-8249
  • Martínez-Bisbal, M.Carmen|||0000-0002-8526-4413
  • Navarro., Clara
  • Vicente Robledo, Javier
  • Tortajada Velert, Salvador
  • Calvar ., Jorge
  • Griffiths ., John
  • Julia Sape, Margarita
  • Howe ., Franklyn A
  • Pujol ., Jesús
  • Peet ., Andrew C
  • Heerschap ., Arend
  • Moreno Torres, Àngel
  • Martinez Granados, Beatriz
  • Wesseling ., Pieter
  • Semmler ., Wolfhard
  • Capellades ., Jaume
  • Majós ., Carles
  • Alberich Bayarri, Ángel
  • Capdevila ., Antoni
  • Monleón ., Daniel
  • Marti Bonmati, Luis
  • Arús ., Carles
  • Celda ., Bernardo
  • Robles Viejo, Montserrat
Object: This study demonstrates that 3T SV-MRS data can be used with the currently available automatic brain tumour diagnostic classifiers which were trained on databases of 1.5T spectra. This will allow the existing large databases of 1.5T MRS data to be used for diagnostic classification of 3T spectra, and perhaps also the combination of 1.5T and 3T databases. Materials and methods: Brain tumour classifiers trained with 154 1.5T spectra to discriminate among high grade malignant tumours and common grade II glial tumours were evaluated with a subsequently-acquired set of 155 1.5T and 37 3T spectra. A similarity study between spectra and main brain tumour metabolite ratios for both field strengths (1.5T and 3T) was also performed. Results: Our results showed that classifiers trained with 1.5T samples had similar accuracy for both test datasets (0.87 ± 0.03 for 1.5T and 0.88 ± 0.03 for 3.0T). Moreover, non-significant differences were observed with most metabolite ratios and spectral patterns. Conclusion: These results encourage the use of existing classifiers based on 1.5T datasets for diagnosis with 3T 1H SV-MRS. The large 1.5T databases compiled throughout many years and the prediction models based on 1.5T acquisitions can therefore continue to be used with data from the new 3T instruments. © 2011 ESMRMB., We would like to thank Miriam Camison-Sanchez for help in the quality control of the MRS data and diagnosis validation. This work was partially funded by the European Commission (FP6-2002-LIFESCIHEALTH 503094) and (IST-2004-27214), the I + D support program of the Universitat Politecnica de Valencia and by the Health Institute Carlos III through the RETICS Combiomed, RD07/0067/2001. CIBER-BBN is an initiative funded by the VI National R&D&D&I Plan 2008-2011, CIBER Actions and financed by the Instituto de Salud Carlos III with assistance from the European Regional Development Fund. The authors acknowledge to Programa Torres Quevedo from Ministerio de Educacion y Ciencia, co-founded by the European Social Fund (PTQ05-02-03386, PTQ-08-01-06802, PTQ-08-01-06817).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Use of 1H and 31P HRMAS to evaluate the relationship between quantitative alterations in metabolite concentrations and tissue features in human brain tumour biopsies

  • Esteve Moya, Vicent
  • Celda,Bernardo
  • Martínez-Bisbal, M.Carmen|||0000-0002-8526-4413
[EN] Quantitative multinuclear high-resolution magic angle spinning (HRMAS) was performed in order to determine the tissue pH values of and the absolute metabolite concentrations in 33 samples of human brain tumour tissue. Metabolite concentrations were quantified by 1D 1 H and 31P HRMAS using the electronic reference to in vivo concentrations (ERETIC) synthetic signal. 1 H–1 H homonuclear and 1 H–31P heteronuclear correlation experiments enabled the direct assessment of the 1 H–31P spin systems for signals that suffered from overlapping in the 1D 1 H spectra, and linked the information present in the 1D 1 H and 31P spectra. Afterwards, the main histological features were determined, and high heterogeneity in the tumour content, necrotic content and nonaffected tissue content was observed. The metabolite profiles obtained by HRMAS showed characteristics typical of tumour tissues: rather low levels of energetic molecules and increased concentrations of protective metabolites. Nevertheless, these characteristics were more strongly correlated with the total amount of living tissue than with the tumour cell contents of the samples alone, which could indicate that the sampling conditions make a significant contribution aside from the effect of tumour development in vivo. The use of methylene diphosphonic acid as a chemical shift and concentration reference for the 31P HRMAS spectra of tissues presented important drawbacks due to its interaction with the tissue. Moreover, the pH data obtained from 31P HRMAS enabled us to establish a correlation between the pH and the distance between the N(CH3)3 signals of phosphocholine and choline in 1 H spectra of the tissue in these tumour samples., The authors acknowledge the SCSIE-University of Valencia Microscopy Service for the histological preparations. They also acknowledge Martial Piotto (Bruker BioSpin, France) for providing the ERETIC synthetic signal. Furthermore, they acknowledge financial support from the Spanish Government project SAF2007-6547, the Generalitat Valenciana project GVACOMP2009-303, and the E.U.'s VI Framework Programme via the project "Web accessible MR decision support system for brain tumor diagnosis and prognosis, incorporating in vivo and ex vivo genomic and metabolomic data" (FP6-2002-LSH 503094). CIBER-BBN is an initiative funded by the VI National R&D&D&i Plan 2008-2011, Iniciativa Ingenio 2010, Consolider Program, CIBER Actions, and financed by the Instituto de Salud Carlos III with assistance from the European Regional Development Fund.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Estudio del colículo inferior de pacientes con esquizofrenia mediante espectroscopía de resonancia magnética, Study of the inferior colliculus in patients with schizophrenia by magnetic resonance spectroscopy

  • Martínez-Granados, Beatriz
  • Sanjuán, Julio
  • Aguilar, Eduardo J.
  • Martí-Bonmatí, Luis
  • Molla, Enrique
  • Celda, Bernardo
  • Martínez-Bisbal, M.Carmen|||0000-0002-8526-4413
[ES] Introducción. Algunos estudios anteriores en pacientes con esquizofrenia han sugerido alteraciones morfométricas y funcionales en el colículo inferior. Las alucinaciones auditivas son uno de los síntomas centrales en la esquizofrenia. Se piensa que en este evento complejo y multidisciplinar, tanto la atención como la emoción desempeñan un papel clave. Objetivo. Estudiar los cambios metabólicos en el colículo inferior, un núcleo integrado en la vía auditiva, en pacientes con esquizofrenia y su posible relación con las alucinaciones auditivas. Sujetos y métodos. Se llevaron a cabo estudios de espectroscopía de resonancia magnética en 30 pacientes diestros con esquizofrenia crónica (19 de ellos con alucinaciones auditivas) y 28 controles. Se adquirió una secuencia 2D de espectroscopía de resonancia magnética y se seleccionaron los vóxeles representativos de ambos colículos inferiores. Se calculó el área de los picos de N-acetilaspartato (NAA), creatina (Cr) y colina (Co). Resultados. Los pacientes con esquizofrenia mostraron una reducción significativa de NAA/Cr en el colículo inferior derecho comparados con los sujetos control. Los datos metabólicos en el colículo inferior derecho se correlacionaron con los ítems emocionales de las alucinaciones auditivas. Conclusiones. La contribución del colículo inferior a las bases neuronales de las alucinaciones auditivas es particularmente relevante para el colículo inferior derecho y se centra en el componente atencional-emocional de este síntoma., [EN] Introduction. Previous studies have suggested morphometric and functional abnormalities in the inferior colliculus in patients with schizophrenia. Auditory hallucinations are one of the central symptoms in schizophrenia. In this complex and multidimensional event both attention and emotion are thought to play a key role. Aim. To study metabolic changes in the inferior colliculus, a nucleus integrated in the auditory pathway, in patients with schizophrenia and the possible relationship with auditory hallucinations. Subjects and methods. Magnetic resonance spectroscopic imaging studies were performed in 30 right-handed patients with chronic schizophrenia (19 of them with auditory hallucinations) and 28 controls. A magnetic resonance spectroscopic imaging 2D slice was acquired and the voxels representative of both inferior colliculi were selected. N-acetylaspartate (NAA), creatine (Cr) and choline (Cho) peak areas were measured. Results. The patients with schizophrenia showed a NAA/Cr significant reduction in the right inferior colliculus compared to the control subjects. The metabolic data in the right inferior colliculus were correlated with emotional auditory hallucinations items. Conclusions. The contribution of the inferior colliculus on neural underpinnings of auditory hallucinations is particularly relevant for the right inferior colliculus and is centered on attention-emotional component of this symptom., Subvención española SAF2007- 65473, Proyecto Europeo Integrado eTUMOUR (FP6-2002-LIFESCIHEALTH 503094) y proyecto MIND del CENIT.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Extracting MRS discriminant functional features of brain tumors

  • Fuster García, Elíes|||0000-0002-0716-8960
  • Garcia-Gomez, Juan M
  • Tortajada Velert, Salvador
  • Vicente Robledo, Javier
  • Robles Viejo, Montserrat
The current challenge in automatic brain tumor classification based on MRS is the improvement of the robustness of the classification models that explicitly account for the probable breach of the independent and identically distributed conditions in the MRS data points. To contribute to this purpose, a new algorithm for the extraction of discriminant MRS features of brain tumors based on a functional approach is presented. Functional data analysis based on region segmentation (RSFDA) is based on the functional data analysis formalism using nonuniformly distributed B splines according to spectral regions that are highly correlated. An exhaustive characterization of the method is presented in this work using controlled and real scenarios. The performance of RSFDA was compared with other widely used feature extraction methods. In all simulated conditions, RSFDA was proven to be stable with respect to the number of variables selected and with respect to the classification performance against noise and baseline artifacts. Furthermore, with real multicenter datasets classification, RSFDA and peak integration (PI) obtained better performance than the other feature extraction methods used for comparison. Other advantages of the method proposed are its usefulness in selecting the optimal number of features for classification and its simplified functional representation of the spectra, which contributes to highlight the discriminative regions of the MR spectrum for each classification task. © 2012 John Wiley & Sons, Ltd., The authors gratefully acknowledge former INTERPRET and eTUMOUR European project partners. Data providers: Professor B. Celda (Physical Chemistry, University of Valencia, Burjassot, Valencia, Spain); Dr F. A. Howe (St George's University of London, London, UK); Dr D. Monleon (Fundacion Investigacion HCUV-University of Valencia, Valencia, Spain); Professor A. Heerschap (Radboud University, Nijmegen, the Netherlands.); Dr. W. Gajewicz, Professor L. Stefanczyk and Dr J. Fortuniak (Uniwersytet Medycznyw Lodz, Lodz, Poland); Professor J. Griffiths (CR UK Cambridge Research Institute, Cambridge, UK); Professor A. C. Peet (Academic Department of Paediatrics and Child Health, University of Birmingham, Birmingham, UK); Professor W. Semmler (Department of Medical Physics in Radiology, German Cancer Research Center, Heidelberg, Germany); Dr. J. Calvar (Fundacion para la Lucha contra las Enfermedades Neurologicas de la Infancia, Buenos Aires, Argentina); Dr. J. Capellades (Hospital Universitari Germans Trias i Pujol, Badalona, Spain); and Dr. C. Majos (Hospital Universitari de Bellvitge, L'Hospitalet de Llobregat, Barcelona, Spain). Data curators: Professor C. Arus, Dr M. Julia-Sape, Dr A. P. Candiota, Dr I. Olier, Ms T. Delgado, Ms J. Martin, Ms M. Camison and Mr A. Perez [all from Grup d'Aplicacions Biomediques de la Ressonancia Magestica Nuclear (GABRMN), Universitat Aut noma de Barcelona (UAB) (GABRMN-UAB) and CIBER-BBN]; and Professor B. Celda, Dra. M. C. Martinez-Bisbal and Dra. B. Martinez-Granados (all from Physical Chemistry, University of Valencia, Burjassot, Valencia, Spain). The authors would also like to thank Dr F. A. Howe and F. Raschke for their suggestions and comments. This work was partially funded by the European Commission: eTUMOUR (contract no. FP6-2002-LIFESCIHEALTH 503094), the HEALTHAGENTS EC project (HEALTHAGENTS) (contract no. FP6-2005-IST 027213).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Magnetic resonance microscopy contribution to interpret High-Resolution Magic Angle Spinning Metabolomic Data of Human Tumor Tissue

  • Martínez-Bisbal, M.Carmen
  • Esteve Moya, Vicent
  • Martinez Granados, Beatriz
  • Celda, Bernardo
[EN] HRMAS NMR is considered a valuable technique to obtain detailed metabolic profile of unprocessed tissues. To properly interpret the HRMAS metabolomic results, detailed information of the actual state of the sample inside the rotor is needed. MRM (Magnetic Resonance Microscopy) was applied for obtaining structural and spatially localized metabolic information of the samples inside the HRMAS rotors. The tissue was observed stuck to the rotor wall under the effect of HRMAS spinning. MRM spectroscopy showed a transference of metabolites from the tissue to the medium. The sample shape and the metabolite transfer after HRMAS indicated that tissue had undergone alterations and it can not be strictly considered as intact. This must be considered when HRMAS is used for metabolic tissue characterization, and it is expected to be highly dependent on the manipulation of the sample. The localized spectroscopic information of MRM reveals the biochemical compartmentalization on tissue samples hidden in the HRMAS spectrum., The authors are grateful for the financial support of Spanish Government SAF2007-6547 and VI Framework E.U. FP6-2002-LSH 503094’ projects.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research

  • Conesa, Ana
  • Götz, Stefan
  • Terol, Javier
  • Talón, Manuel
  • Robles Viejo, Montserrat
  • Garcia-Gomez, Juan M
[EN] We present here Blast2GO (B2G), a research tool designed with the main purpose of enabling Gene Ontology (GO) based data mining on sequence data for which no GO annotation is yet available. B2G joints in one application GO annotation based on similarity searches with statistical analysis and highlighted visualization on directed acyclic graphs. This tool offers a suitable platform for functional genomics research in non-model species. B2G is an intuitive and interactive desktop application that allows monitoring and comprehension of the whole annotation and analysis process., The authors thank Dr Timothy Williams for fruitful discussions andcomments on the software and Nils Bluethgen for kindly providingthe Gossip software and supporting integration in B2G. This workhas been funded by MCyT (GEN 2001-4885-C05-03) and eTumourProject (FP6-2002-LIFESCIHEALTH 503094). The authors thankthe INBIOMED G03/160 research thematic network financed byFIS of the Instituto de Salud Carlos III.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Effect of feature extraction for brain tumor classification based on short echo time 1H MR spectra

  • Luts, Jan
  • Poullet, Jean-Baptiste
  • Heerschap, Arend
  • Robles, Montserrat
  • Suykens, Johan A.K.
  • Van Huffel, Sabine
  • Garcia-Gomez, Juan M
[EN] This study examines the effect of feature extraction methods prior to automated pattern recognition based on magnetic resonance spectroscopy (MRS) for brain tumor diagnosis. Since individual inspection of spectra is time-consuming and requires specific spectroscopic expertise, the introduction of clinical decision support systems.(DSSs) is expected to strongly promote the clinical use of MRS. This study focuses on the feature extraction step in the preprocessing protocol of MRS when using a DSS. On two independent data sets, encompassing single-voxel and multi-voxel data, it is observed that the use of the full spectra together with a kernel-based technique, handling high dimensional data, or using an automated pattern recognition method based on independent component analysis or Relief-F achieves accurate performances. In addition, these approaches have low cost and are easy to automate. When sophisticated quantification methods are used in a DSS, user interaction should be minimized. The computationally intensive quantification techniques do not tend to increase the performance in these circumstances. The results suggest to simplify the feature reduction step in the preprocessing protocol when using a DSS purely for classification purposes. This can greatly speed up the execution of classifiers and DSSs and may accelerate their introduction into clinical practice., The authors thank the Institute for Molecules and Materials, Analytical Chemistry, Chemometrics Research Department of the Radboud University Nijmegen for preprocessing the MRSI data. Margarida Julia-Sape is gratefully acknowledged for all data management tasks; the clinical partners represented by Dr. Witold Gajewicz (MUL, Lodzi, Poland) and Jorge Andres Calvar (FLENI, Ciudad de Buenos Aires, Argentina) are gratefully acknowledged for providing data. Grant sponsors: Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen), Research Council KUL: GOA-AMBioRICS, Centers-of-Excellence in Optimisation; Grant sponsor: IDO; Grant number: 05/010 EEG-fMRI; Grant sponsor: Flemish Government: FWO; Grant numbers: G.0360.05, G.0519.06, G.0341.07, G.0321.06, G.0302.07; Grant sponsors: ICCoS, ANMMM, IWT; Grant sponsor: Belgian Federal Government: DWTC; Grant numbers: IUAP IV-02 (1996 2001), IUAP V-22 (2002 2006);Grant sponsor: Belgian Federal Science Policy Office;Grant number: IUAP P6/04; Grant sponsor: EU; Grant number: IST-1999-10310; Grant sponsor: BIOPATTERN; Grant number: FP6-2002-IST 508803; Grant sponsor: eTUMOUR; Grant number: FP6-2002-LIFESCIHEALTH 503094; Grant sponsor: HealthAgents; Grant number: FP6-2005-IST 027213; Grant sponsor: FAST; Grant number: FP6-019279-2; Grant sponsor: ESA: Cardiovascular Control; Grant number: Prodex-8 C90242.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Automated Brain Tumor biopsy prediction using single-labeling cDNA Microarrays-based gene expression profiling

  • Castells, Xavier
  • Navarro, Alfredo
  • Acebes, Juan José
  • Godino, Óscar
  • Boluda, Susana
  • Barceló, Anna
  • Robles, Montserrat
  • Ariño, Joaquín
  • Arús, Carles
  • Garcia-Gomez, Juan M
[EN] Aims: Gene signatures obtained from microarray experiments may be of use to improve the prediction of brain tumor diagnosis. Nevertheless, automated and objective prediction with accuracy comparable to or better than the gold standard should be convincingly demonstrated for possible clinician uptake of the new methodology. Herewith, we demonstrate that primary brain tumor types can be discriminated using microarray data in an automated and objective way. Methods: Postsurgical biopsies from 35 patients [17 glioblastoma multiforme (Gbm) and 18 meningothelial meningioma (Mm)] were stored in liquid nitrogen, total RNA was extracted, and cDNA was labeled with Cy3 fluorochrome and hybridized onto a cDNA-based microarray containing 11,500 cDNA clones representing 9300 loci. Scanned data were preprocessed, normalized, and used for predictor development. The predictive functions were fitted to a subset of samples and their performance evaluated with an independent subset. Expression results were validated by means of real time-polymerase chain reaction. Results: Some gene expression-based predictors achieved 100% accuracy both in training resampling validation and independent testing. One of them, composed of GFAP, PTPRZ1, GPM6B and PRELP, produced a 100% prediction accuracy for both training and independent test datasets. Furthermore, the gene signatures obtained, increased cell detoxification, motility and intracellular transport in Gbm, and increased cell adhesion and cytochrome-family genes in Mm, agree well with the expected biologic and pathologic characteristics of the studied tumors. Conclusions: The ability of gene signatures to automate prediction of brain tumors through a fully objective approach has been demonstrated. A comparison of gene expression profiles between Gbm and Mm may provide additional clues about patterns associated with each tumor type., Supported in part by grants from the European Commission (FP6-2002-LIFESCIHEALTH 503094 and IST-2004-27214), the research project MEDIVO2 (MEC, SAF2005-03650), and the Programa de Apoyo a la lnvestigacion y Desarrollo (PAID-00-06 UPV)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Magnetic resonance spectroscopy and imaging on fresh human brain tumor biopsies at microscopic resolution

  • Martínez-Bisbal, M.Carmen
  • Martínez-Granados, Beatriz
  • Rovira, Vicente
  • Celda, Bernardo
  • Esteve, Vicent
[EN] The metabolic compn. and concn. knowledge provided by magnetic resonance spectroscopy (MRS) liq. and high-resoln. magic angle spinning spectroscopy (HR-MAS) has a relevant impact in clin. practice during magnetic resonance imaging (MRI) monitoring of human tumors. In addn., the combination of morphol. and chem. information by MRI and MRS has been particularly useful for diagnosis and prognosis of tumor evolution. MRI spatial resoln. reachable in human beings is limited for safety reasons and the demanding necessary conditions are only applicable on exptl. model animals. Nevertheless, MRS and MRI can be performed on human biopsies at high spatial resoln., enough to allow a direct correlation between the chem. information and the histol. features obsd. in such biopsies. Although HR-MAS is nowadays a well-established technique for spectroscopic anal. of tumor biopsies, with this approach just a mean metabolic profile of the whole sample can be obtained and thus the high histol. heterogeneity of some important tumors is mostly neglected. The value of metabolic HR-MAS data strongly depends on a wide statistical anal. and usually the microanatomical rationale for the correlation between histol. and spectroscopy is lost. We present here a different approach for the combined use of MRI and MRS on fresh human brain tumor biopsies with native contrast. This approach has been designed to achieve high spatial (18 × 18 × 50 m) and spectral (0.031 L) resoln. in order to obtain as much spatially detailed morphol. and metabolical information as possible without any previous treatment that can alter the sample. The preservation of native tissue conditions can provide information that can be translated to in vivo studies and addnl. opens the possibility of performing other techniques to obtain complementary information from the same sample., The authors acknowledge the SCSIE-University of Valencia Microscopy Service for the histological preparations. They also acknowledge financial support from the Spanish Government project SAF2007-6547, the Generalitat Valenciana project GVACOMP2009-303, and the E.U.s VI Framework Program via the project "Web accessible MR decision support system for brain tumor diagnosis and prognosis, incorporating in vivo and ex vivo genomic and metabolomic data" (FP6-2002-LSH 503094).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Accurate classification of childhood brain tumours by in vivo 1H MRS - a multi-centre study

  • Vicente Robledo, Javier
  • Tortajada Velert, Salvador
  • Davies, Nigel
  • Natarajan, Kal
  • Wilson, Martin
  • Grundy, Richard G.
  • Wesseling, Pieter
  • Monleón, Daniel
  • Celda, Bernardo
  • Robles, Montserrat
  • Peet, Andrew C.
  • Fuster García, Elíes|||0000-0002-0716-8960
  • Garcia-Gomez, Juan M
Aims: To evaluate the accuracy of single-voxel Magnetic Resonance Spectroscopy (1H-MRS) as a non-invasive diagnostic aid for pediatric brain tumours in a multi-national study. Our hypotheses are (1) that automated classification based on 1H-MRS provides an accurate non-invasive diagnosis in multi-centre datasets and (2) using a protocol which increases the metabolite information improves the diagnostic accuracy. Methods: 78 patients under 16 years old with histologically proven brain tumours from 10 international centres were investigated. Discrimination of 29 medulloblastomas, 11 ependymomas and 38 pilocytic astrocytomas was evaluated. Single-voxel MRS was undertaken prior to diagnosis (1.5Tesla PRESS, PROBE or STEAM, TE 20-32 ms, and 135-136 ms). MRS data was processed using two strategies, determination of metabolite concentrations using TARQUIN software and automatic feature extraction with Peak Integration. Linear Discriminant Analysis was applied to this data to produce diagnostic classifiers. An evaluation of the diagnostic accuracy was performed based on resampling to measure the Balanced Accuracy Rate (BAR). Results: The accuracy of the diagnostic classifiers for discriminating the three tumour types was found to be high (BAR 0.98) when a combination of TE was used. The combination of both TE significantly improved the classification performance (p < 0.01, Tukey¿s test) compared with the use of one TE alone. 3 Other tumour types were classified accurately as glial or primitive neuroectodermal (BAR 1.00). Conclusions: 1H-MRS has excellent accuracy for the non-invasive diagnosis of common childhood brain tumours particularly if the metabolite information is maximised and should become part of routine clinical assessment for these children., This work was funded by the European Commission (FP6-2002-LIFESCIHEALTH 503094). Additional analysis was made available through the CR UK and EPSRC Cancer Imaging Programme at the Children's cancer and Leukaemia Group in association with the MRC and Department of Health (England) (C7809/A10342). We thank eTUMOUR partners for providing data, in particular J. Capellades (IDI-Badalona), C. Majos (IDI-Bellvitge), A. Moreno (Centre Diagnostic Pedralbes), J. Calvar (FLENI) and A. Capdevila (H. Sant Joan de Deu).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

The effect of combining two echo times in automatic brain tumor classification by MRS

  • Garcia-Gomez, Juan M
  • Tortajada, Salvador
  • Vidal, César
  • Julià -Sapé, Margalida
  • Luts, Jan
  • Moreno-Torres, Àngel
  • Van Huffel, Sabine
  • Arús, Carles
  • Robles Viejo, Montserrat
[EN] H-1 MRS is becoming an accurate, non-invasive technique for initial examination of brain masses. We investigated if the combination of single-voxel H-1 MRS at 1.5 T at two different (TEs), short TE (PRESS or STEAM, 20-32 ms) and long TE (PRESS, 135-136 ms), improves the classification of brain tumors over using only one echo TE. A clinically validated dataset of 50 low-grade meningiomas, 105 aggressive tumors (glioblastoma and metastasis). and 30 low-grade glial tumors (astrocytomas grade II, oligodendrogliomas anti oligoastrocytomas) was used to fit predictive models based on the combination of features from short-TEs and long-TE spectra. A new approach that combines the two consecutively was used to produce a single data vector from which relevant features of the two TE spectra could be extracted by means of three algorithms: stepwise, reliefF, and principal components analysis. Least Squares support vector machines and linear discriminant analysis were applied to fit the pairwise and multiclass classifiers, respectively. Significant differences in performance were found when short-TE, long-TE or both spectra combined were used as input. In our dataset, to discriminate meningiomas, the combination of the two TE acquisitions produced optimal performance. To discriminate aggressive tumors from low-grade filial tumours, the use of short-TE acquisition alone was preferable. The classifier development strategy used here lends itself to automated learning and test performance processes, which may, be of use for future web-based multicentric classifier development studies. Copyright (c) 2008 John Wiley & Sons, Ltd., This work was partially funded by the European Commission: eTUMOUR (contract no. FP6-2002-LIFESCI-HEALTH 503094). HealthAgents (contract no. FP6-2005-IST 027213), BIOPATTERN (contract no. FP6-2002-IST 508803); Programa de Apoyo a la Investigacion y Desarrollo, PAID-00-06 UPV; Research Council KUL: GOA-AMBioRICS, Centers-of-excellence optimisation; Belgian Federal Government: DWTC, IUAPV P6/04 (DYSCO 2007-2011). Jan Luts is a PhD student supported by an IWT grant. We thank INTERPRET partners for providing data, in particular, C. Majos (IDI-Bellvitge), John Griffiths (SGUL), Arend Heerschap (RU), Witold Gajewicz (MUL), and Jorge Calvar (FLENI). We also thank Guillem Mercadal for his feedback about 1H MRS data processing.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Multiproject-multicenter evaluation of automatic brain tumor classification by magnetic resonance spectroscopy

  • Garcia-Gomez, Juan M
  • Fuster García, Elíes|||0000-0002-0716-8960
  • Martínez-Bisbal, M.Carmen|||0000-0002-8526-4413
  • Luts, Jan
  • Julia-Sape, Margarida
  • Krooshof, Patrick
  • Tortajada Velert, Salvador
  • Vicente Robledo, Javier
  • Melssen, Willem
  • Olier, Ivan
  • Postma, Geert
  • Monleon, Daniel
  • Moreno-Torres, Angel
  • Pujol, Jesus
  • Candiota, Ana-Paula
  • Suykens, Johan
  • Buydens, Lutgarde
  • Celda, Bernardo
  • Van Huffel, Sabine
  • Arus, Carles
  • Robles Viejo, Montserrat
[EN] Automatic brain tumor classification by MRS has been under development for more than a decade. Nonetheless, to our knowledge, there are no published evaluations of predictive models with unseen cases that are subsequently acquired in different centers. The multicenter eTUMOUR project (2004-2009), which builds upon previous expertise from the INTERPRET project (2000-2002) has allowed such an evaluation to take place. A total of 253 pairwise classifiers for glioblastoma, meningioma, metastasis, and low-grade glial diagnosis were inferred based on 211 SV short TE INTERPRET MR spectra obtained at 1.5 T (PRESS or STEAM, 20-32 ms) and automatically pre-processed. Afterwards, the classifiers were tested with 97 spectra, which were subsequently compiled during eTUMOUR. In our results based on subsequently acquired spectra, accuracies of around 90% were achieved for most of the pairwise discrimination problems. The exception was for the glioblastoma versus metastasis discrimination, which was below 78%. A more clear definition of metastases may be obtained by other approaches, such as MRSI + MRI. The prediction of the tumor type of in-vivo MRS is possible using classifiers developed from previously acquired data, in different hospitals with different instrumentation under the same acquisition protocols. This methodology may find application for assisting in the diagnosis of new brain tumor cases and for the quality control of multicenter MRS databases., We would like to thank the INTERPRET and eTUMOUR partners for providing data, particularly, Carles Majos (IDI-Bellvitge), John Griffiths and Franklyn Howe (SGUL), Arend Heerschap (RU), Witold Gajewicz (MUL), Jorge Calvar (FLENI), and Antoni Capdevila (H. de Sant Joan de Deu). This work was partially funded by the European Commission: eTUMOUR (contract no. FP62002-LIFESCIHEALTH 503094), the HEALTHAGENTS EC project (HEALTHAGENTS) (contract no. FP6-2005-IST 027213), BIOPATTERN (contract no. FP6-2002-IST 508803). The authors appreciate the suggestions from the reviewers that have improved the discussion presented in this work. We also thank the following for their contributions: Programa de Apoyo a la Investigacion y Desarrollo, PAID-00-06 UPV; Research Council KUL: GOA-AMBioRICS, Centers-of-excellence optimisation; Belgian Federal Government: DWTC, IUAPV P6/04 (DYSCO 2007-2011); the following participants acknowledge the following: JVR acknowledges to Programa Torres Quevedo from Ministerio de Educacion y Ciencia, co-founded by the European Social Fund (PTQ05-02-03386). JL is a PhD student supported by an IWT grant. DM is supported by the Ministerio de Educacion y Ciencia del Gobierno de Espa a for a Ramon y Cajal 2006 Contract. BC and CA gratefully acknowledge the Ministerio de Educacion y Ciencia del Gobierno de Espa a (BC: SAF2004-06297 and SAF2007-6547; CA: SAF2005-03650). CIBER-BBN is an initiative of the "Instituto de Salud Carlos III" (ISCIII), Spain
Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

Development of robust discriminant equations for assessing subtypes of glioblastoma biopsies

  • Castells Domingo, Xavier
  • Acebes Martín, Juan José
  • Majos, Carles|||0000-0003-0468-5150
  • Boluda, Susana
  • Julià Sapé, Ma. Margarita|||0000-0002-3316-9027
  • Candiota Silveira, Ana Paula|||0000-0002-1523-6505
  • Ariño Carmona, Joaquín|||0000-0002-6774-2987
  • Barceló, Antonia
  • Arús i Caraltó, Carles|||0000-0003-2510-2671
In the preceding decade, various studies on glioblastoma (Gb) demonstrated that signatures obtained from gene expression microarrays correlate better with survival than with histopathological classification. However, there is not a universal consensus formula to predict patient survival. We developed a gene signature using the expression profile of 47 Gbs through an unsupervised procedure and two groups were obtained. Subsequent to a training procedure through leave-one-out cross-validation, we fitted a discriminant (linear discriminant analysis (LDA)) equation using the four most discriminant probesets. This was repeated for two other published signatures and the performance of LDA equations was evaluated on an independent test set, which contained status of IDH1 mutation, EGFR amplification, MGMT methylation and gene VEGF expression, among other clinical and molecular information. The unsupervised local signature was composed of 69 probesets and clearly defined two Gb groups, which would agree with primary and secondary Gbs. This hypothesis was confirmed by predicting cases from the independent data set using the equations developed by us. The high survival group predicted by equations based on our local and one of the published signatures contained a significantly higher percentage of cases displaying IDH1 mutation and non-amplification of EGFR. In contrast, only the equation based on the published signature showed in the poor survival group a significant high percentage of cases displaying a hypothesised methylation of MGMT gene promoter and overexpression of gene VEGF. We have produced a robust equation to confidently discriminate Gb subtypes based in the normalised expression level of only four genes.