Performance comparison of modified ComBat for harmonization of radiomic features for multicenter studies View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2020-06-24

AUTHORS

R. Da-ano, I. Masson, F. Lucia, M. Doré, P. Robin, J. Alfieri, C. Rousseau, A. Mervoyer, C. Reinhold, J. Castelli, R. De Crevoisier, J. F. Rameé, O. Pradier, U. Schick, D. Visvikis, M. Hatt

ABSTRACT

Multicenter studies are needed to demonstrate the clinical potential value of radiomics as a prognostic tool. However, variability in scanner models, acquisition protocols and reconstruction settings are unavoidable and radiomic features are notoriously sensitive to these factors, which hinders pooling them in a statistical analysis. A statistical harmonization method called ComBat was developed to deal with the “batch effect” in gene expression microarray data and was used in radiomics studies to deal with the “center-effect”. Our goal was to evaluate modifications in ComBat allowing for more flexibility in choosing a reference and improving robustness of the estimation. Two modified ComBat versions were evaluated: M-ComBat allows to transform all features distributions to a chosen reference, instead of the overall mean, providing more flexibility. B-ComBat adds bootstrap and Monte Carlo for improved robustness in the estimation. BM-ComBat combines both modifications. The four versions were compared regarding their ability to harmonize features in a multicenter context in two different clinical datasets. The first contains 119 locally advanced cervical cancer patients from 3 centers, with magnetic resonance imaging and positron emission tomography imaging. In that case ComBat was applied with 3 labels corresponding to each center. The second one contains 98 locally advanced laryngeal cancer patients from 5 centers with contrast-enhanced computed tomography. In that specific case, because imaging settings were highly heterogeneous even within each of the five centers, unsupervised clustering was used to determine two labels for applying ComBat. The impact of each harmonization was evaluated through three different machine learning pipelines for the modelling step in predicting the clinical outcomes, across two performance metrics (balanced accuracy and Matthews correlation coefficient). Before harmonization, almost all radiomic features had significantly different distributions between labels. These differences were successfully removed with all ComBat versions. The predictive ability of the radiomic models was always improved with harmonization and the improved ComBat provided the best results. This was observed consistently in both datasets, through all machine learning pipelines and performance metrics. The proposed modifications allow for more flexibility and robustness in the estimation. They also slightly but consistently improve the predictive power of resulting radiomic models. More... »

PAGES

10248

References to SciGraph publications

  • 2006-02-23. Bias in error estimation when using cross-validation for model selection in BMC BIOINFORMATICS
  • 2010-07-30. A comparison of batch effect removal methods for enhancement of prediction performance using MAQC-II microarray gene expression data in THE PHARMACOGENOMICS JOURNAL
  • 2009. The Elements of Statistical Learning, Data Mining, Inference, and Prediction in NONE
  • 2017-12-08. Ten quick tips for machine learning in computational biology in BIODATA MINING
  • 2018-03-21. Multiregional radiomics features from multiparametric MRI for prediction of MGMT methylation status in glioblastoma multiforme: A multicentre study in EUROPEAN RADIOLOGY
  • 2018-03-02. Feasibility of state of the art PET/CT systems performance harmonisation in EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING
  • 1995. The Nature of Statistical Learning Theory in NONE
  • 2006-01-01. Embedded Methods in FEATURE EXTRACTION
  • 2017-12-09. Prediction of outcome using pretreatment 18F-FDG PET/CT and MRI radiomics in locally advanced cervical cancer treated with chemoradiotherapy in EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING
  • 2018-12-07. External validation of a combined PET and MRI radiomics model for prediction of recurrence in cervical cancer patients treated with chemoradiotherapy in EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING
  • 2018-01-11. Caveolin 1 Promotes Renal Water and Salt Reabsorption in SCIENTIFIC REPORTS
  • 2001-10. Random Forests in MACHINE LEARNING
  • 2012-06-08. Batch correction of microarray data substantially improves the identification of genes differentially expressed in Rheumatoid Arthritis and Osteoarthritis in BMC MEDICAL GENOMICS
  • 2017-10-04. Radiomics: the bridge between medical imaging and personalized medicine in NATURE REVIEWS CLINICAL ONCOLOGY
  • 2014-12-02. FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0 in EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1038/s41598-020-66110-w

    DOI

    http://dx.doi.org/10.1038/s41598-020-66110-w

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1128717081

    PUBMED

    https://www.ncbi.nlm.nih.gov/pubmed/32581221


    Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
    Incoming Citations Browse incoming citations for this publication using opencitations.net

    JSON-LD is the canonical representation for SciGraph data.

    TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

    [
      {
        "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
        "about": [
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Mathematical Sciences", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Information and Computing Sciences", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0104", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Statistics", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Artificial Intelligence and Image Processing", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Female", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Humans", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Laryngeal Neoplasms", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Machine Learning", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Magnetic Resonance Imaging", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Multicenter Studies as Topic", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Positron-Emission Tomography", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Prognosis", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Radiographic Image Interpretation, Computer-Assisted", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Uterine Cervical Neoplasms", 
            "type": "DefinedTerm"
          }
        ], 
        "author": [
          {
            "affiliation": {
              "alternateName": "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France", 
              "id": "http://www.grid.ac/institutes/grid.6289.5", 
              "name": [
                "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Da-ano", 
            "givenName": "R.", 
            "id": "sg:person.011560743623.06", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011560743623.06"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Radiation Oncology, Institut de canc\u00e9rologie de l\u2019Ouest Ren\u00e9-Gauducheau, Saint-Herblain, France", 
              "id": "http://www.grid.ac/institutes/None", 
              "name": [
                "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France", 
                "Department of Radiation Oncology, Institut de canc\u00e9rologie de l\u2019Ouest Ren\u00e9-Gauducheau, Saint-Herblain, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Masson", 
            "givenName": "I.", 
            "id": "sg:person.01233533242.82", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01233533242.82"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Radiation Oncology Department, University Hospital, Brest, France", 
              "id": "http://www.grid.ac/institutes/grid.411766.3", 
              "name": [
                "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France", 
                "Radiation Oncology Department, University Hospital, Brest, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Lucia", 
            "givenName": "F.", 
            "id": "sg:person.07532416715.90", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07532416715.90"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Radiation Oncology, Institut de canc\u00e9rologie de l\u2019Ouest Ren\u00e9-Gauducheau, Saint-Herblain, France", 
              "id": "http://www.grid.ac/institutes/None", 
              "name": [
                "Department of Radiation Oncology, Institut de canc\u00e9rologie de l\u2019Ouest Ren\u00e9-Gauducheau, Saint-Herblain, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Dor\u00e9", 
            "givenName": "M.", 
            "id": "sg:person.0713341730.08", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0713341730.08"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Nuclear Medicine, University of Brest, Brest, France", 
              "id": "http://www.grid.ac/institutes/grid.6289.5", 
              "name": [
                "Department of Nuclear Medicine, University of Brest, Brest, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Robin", 
            "givenName": "P.", 
            "id": "sg:person.01307140060.86", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01307140060.86"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Radiation Oncology, McGill University Health Centre, Montreal, Quebec, Canada", 
              "id": "http://www.grid.ac/institutes/grid.63984.30", 
              "name": [
                "Department of Radiation Oncology, McGill University Health Centre, Montreal, Quebec, Canada"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Alfieri", 
            "givenName": "J.", 
            "id": "sg:person.01046015545.10", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01046015545.10"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "CRCINA, University of Nantes, INSERM UMR1232, CNRS-ERL6001, Nantes, France", 
              "id": "http://www.grid.ac/institutes/grid.4817.a", 
              "name": [
                "Department of Nuclear Medicine, Institut de cancerologie de l\u2019Ouest Ren\u00e9-Gauducheau, Saint-Herblain, France", 
                "CRCINA, University of Nantes, INSERM UMR1232, CNRS-ERL6001, Nantes, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Rousseau", 
            "givenName": "C.", 
            "id": "sg:person.01240530667.47", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01240530667.47"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Radiation Oncology, Institut de canc\u00e9rologie de l\u2019Ouest Ren\u00e9-Gauducheau, Saint-Herblain, France", 
              "id": "http://www.grid.ac/institutes/None", 
              "name": [
                "Department of Radiation Oncology, Institut de canc\u00e9rologie de l\u2019Ouest Ren\u00e9-Gauducheau, Saint-Herblain, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Mervoyer", 
            "givenName": "A.", 
            "id": "sg:person.0776411042.55", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0776411042.55"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Radiology, McGill University Health Centre, Montreal, Canada", 
              "id": "http://www.grid.ac/institutes/grid.63984.30", 
              "name": [
                "Department of Radiology, McGill University Health Centre, Montreal, Canada"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Reinhold", 
            "givenName": "C.", 
            "id": "sg:person.01134767703.78", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01134767703.78"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "University of Rennes 1, LTSI, Rennes, France", 
              "id": "http://www.grid.ac/institutes/grid.463996.7", 
              "name": [
                "Radiotherapy Department Cancer, Institute Eugene Marquis, Rennes, France", 
                "University of Rennes 1, LTSI, Rennes, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Castelli", 
            "givenName": "J.", 
            "id": "sg:person.0576143704.48", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0576143704.48"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "University of Rennes 1, LTSI, Rennes, France", 
              "id": "http://www.grid.ac/institutes/grid.463996.7", 
              "name": [
                "Radiotherapy Department Cancer, Institute Eugene Marquis, Rennes, France", 
                "University of Rennes 1, LTSI, Rennes, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "De Crevoisier", 
            "givenName": "R.", 
            "id": "sg:person.01371121274.06", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01371121274.06"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Medical Oncology, Centre Hospitalier de Vendee, La Roche sur Yon, France", 
              "id": "http://www.grid.ac/institutes/None", 
              "name": [
                "Department of Medical Oncology, Centre Hospitalier de Vendee, La Roche sur Yon, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Rame\u00e9", 
            "givenName": "J. F.", 
            "id": "sg:person.010361342027.13", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010361342027.13"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Radiation Oncology Department, University Hospital, Brest, France", 
              "id": "http://www.grid.ac/institutes/grid.411766.3", 
              "name": [
                "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France", 
                "Radiation Oncology Department, University Hospital, Brest, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Pradier", 
            "givenName": "O.", 
            "id": "sg:person.01024223375.85", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01024223375.85"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Radiation Oncology Department, University Hospital, Brest, France", 
              "id": "http://www.grid.ac/institutes/grid.411766.3", 
              "name": [
                "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France", 
                "Radiation Oncology Department, University Hospital, Brest, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Schick", 
            "givenName": "U.", 
            "id": "sg:person.01206441206.87", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01206441206.87"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France", 
              "id": "http://www.grid.ac/institutes/grid.6289.5", 
              "name": [
                "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Visvikis", 
            "givenName": "D.", 
            "id": "sg:person.01255045106.49", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01255045106.49"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France", 
              "id": "http://www.grid.ac/institutes/grid.6289.5", 
              "name": [
                "INSERM, UMR 1101, LaTIM, University of Brest, Brest, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Hatt", 
            "givenName": "M.", 
            "id": "sg:person.01202724075.78", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01202724075.78"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1007/978-1-4757-2440-0", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1027312764", 
              "https://doi.org/10.1007/978-1-4757-2440-0"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s00259-018-3977-4", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1101318626", 
              "https://doi.org/10.1007/s00259-018-3977-4"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1023/a:1010933404324", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1024739340", 
              "https://doi.org/10.1023/a:1010933404324"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1038/s41598-017-19071-6", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1100212558", 
              "https://doi.org/10.1038/s41598-017-19071-6"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1186/s13040-017-0155-3", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1099596880", 
              "https://doi.org/10.1186/s13040-017-0155-3"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s00259-014-2961-x", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1008918527", 
              "https://doi.org/10.1007/s00259-014-2961-x"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1186/1471-2105-7-91", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1034791610", 
              "https://doi.org/10.1186/1471-2105-7-91"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1186/1755-8794-5-23", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1016278320", 
              "https://doi.org/10.1186/1755-8794-5-23"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-0-387-84858-7", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1032573094", 
              "https://doi.org/10.1007/978-0-387-84858-7"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1038/nrclinonc.2017.141", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1092061102", 
              "https://doi.org/10.1038/nrclinonc.2017.141"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-540-35488-8_6", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1031342354", 
              "https://doi.org/10.1007/978-3-540-35488-8_6"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s00259-017-3898-7", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1099696659", 
              "https://doi.org/10.1007/s00259-017-3898-7"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s00259-018-4231-9", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1110448603", 
              "https://doi.org/10.1007/s00259-018-4231-9"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1038/tpj.2010.57", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1030763566", 
              "https://doi.org/10.1038/tpj.2010.57"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s00330-017-5302-1", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1101678461", 
              "https://doi.org/10.1007/s00330-017-5302-1"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2020-06-24", 
        "datePublishedReg": "2020-06-24", 
        "description": "Multicenter studies are needed to demonstrate the clinical potential value of radiomics as a prognostic tool. However, variability in scanner models, acquisition protocols and reconstruction settings are unavoidable and radiomic features are notoriously sensitive to these factors, which hinders pooling them in a statistical analysis. A statistical harmonization method called ComBat was developed to deal with the \u201cbatch effect\u201d in gene expression microarray data and was used in radiomics studies to deal with the \u201ccenter-effect\u201d. Our goal was to evaluate modifications in ComBat allowing for more flexibility in choosing a reference and improving robustness of the estimation. Two modified ComBat versions were evaluated: M-ComBat allows to transform all features distributions to a chosen reference, instead of the overall mean, providing more flexibility. B-ComBat adds bootstrap and Monte Carlo for improved robustness in the estimation. BM-ComBat combines both modifications. The four versions were compared regarding their ability to harmonize features in a multicenter context in two different clinical datasets. The first contains 119 locally advanced cervical cancer patients from 3 centers, with magnetic resonance imaging and positron emission tomography imaging. In that case ComBat was applied with 3 labels corresponding to each center. The second one contains 98 locally advanced laryngeal cancer patients from 5 centers with contrast-enhanced computed tomography. In that specific case, because imaging settings were highly heterogeneous even within each of the five centers, unsupervised clustering was used to determine two labels for applying ComBat. The impact of each harmonization was evaluated through three different machine learning pipelines for the modelling step in predicting the clinical outcomes, across two performance metrics (balanced accuracy and Matthews correlation coefficient). Before harmonization, almost all radiomic features had significantly different distributions between labels. These differences were successfully removed with all ComBat versions. The predictive ability of the radiomic models was always improved with harmonization and the improved ComBat provided the best results. This was observed consistently in both datasets, through all machine learning pipelines and performance metrics. The proposed modifications allow for more flexibility and robustness in the estimation. They also slightly but consistently improve the predictive power of resulting radiomic models.", 
        "genre": "article", 
        "id": "sg:pub.10.1038/s41598-020-66110-w", 
        "isAccessibleForFree": true, 
        "isFundedItemOf": [
          {
            "id": "sg:grant.7070033", 
            "type": "MonetaryGrant"
          }
        ], 
        "isPartOf": [
          {
            "id": "sg:journal.1045337", 
            "issn": [
              "2045-2322"
            ], 
            "name": "Scientific Reports", 
            "publisher": "Springer Nature", 
            "type": "Periodical"
          }, 
          {
            "issueNumber": "1", 
            "type": "PublicationIssue"
          }, 
          {
            "type": "PublicationVolume", 
            "volumeNumber": "10"
          }
        ], 
        "keywords": [
          "multicenter study", 
          "cancer patients", 
          "radiomic features", 
          "advanced cervical cancer patients", 
          "contrast-enhanced computed tomography", 
          "radiomics model", 
          "cervical cancer patients", 
          "laryngeal cancer patients", 
          "magnetic resonance imaging", 
          "positron emission tomography", 
          "clinical outcomes", 
          "prognostic tool", 
          "computed tomography", 
          "multicenter context", 
          "resonance imaging", 
          "emission tomography", 
          "different clinical datasets", 
          "patients", 
          "radiomics studies", 
          "tomography", 
          "acquisition protocols", 
          "clinical datasets", 
          "overall mean", 
          "statistical analysis", 
          "center", 
          "setting", 
          "predictive ability", 
          "expression microarray data", 
          "study", 
          "outcomes", 
          "potential value", 
          "radiomics", 
          "scanner models", 
          "harmonization methods", 
          "reconstruction settings", 
          "gene expression microarray data", 
          "imaging", 
          "predictive power", 
          "microarray data", 
          "combat", 
          "Monte Carlo", 
          "unsupervised clustering", 
          "ability", 
          "better results", 
          "factors", 
          "differences", 
          "features", 
          "performance metrics", 
          "more flexibility", 
          "cases", 
          "protocol", 
          "different distributions", 
          "modelling step", 
          "effect", 
          "modification", 
          "improved robustness", 
          "labels", 
          "estimation", 
          "robustness", 
          "data", 
          "second one", 
          "specific case", 
          "model", 
          "reference", 
          "harmonization", 
          "impact", 
          "performance comparison", 
          "comparison", 
          "variability", 
          "analysis", 
          "Carlo", 
          "batch effects", 
          "results", 
          "version", 
          "bootstrap", 
          "goal", 
          "tool", 
          "values", 
          "means", 
          "metrics", 
          "different machine", 
          "distribution", 
          "method", 
          "feature distribution", 
          "flexibility", 
          "machine", 
          "clustering", 
          "dataset", 
          "step", 
          "context", 
          "one", 
          "power", 
          "pipeline"
        ], 
        "name": "Performance comparison of modified ComBat for harmonization of radiomic features for multicenter studies", 
        "pagination": "10248", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1128717081"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1038/s41598-020-66110-w"
            ]
          }, 
          {
            "name": "pubmed_id", 
            "type": "PropertyValue", 
            "value": [
              "32581221"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1038/s41598-020-66110-w", 
          "https://app.dimensions.ai/details/publication/pub.1128717081"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2022-08-04T17:10", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-springernature-scigraph/baseset/20220804/entities/gbq_results/article/article_866.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://doi.org/10.1038/s41598-020-66110-w"
      }
    ]
     

    Download the RDF metadata as:  json-ld nt turtle xml License info

    HOW TO GET THIS DATA PROGRAMMATICALLY:

    JSON-LD is a popular format for linked data which is fully compatible with JSON.

    curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1038/s41598-020-66110-w'

    N-Triples is a line-based linked data format ideal for batch operations.

    curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1038/s41598-020-66110-w'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1038/s41598-020-66110-w'

    RDF/XML is a standard XML format for linked data.

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1038/s41598-020-66110-w'


     

    This table displays all metadata directly associated to this object as RDF triples.

    393 TRIPLES      21 PREDICATES      145 URIs      120 LITERALS      17 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1038/s41598-020-66110-w schema:about N028e50742e1f48f3be36a3e5608a6772
    2 N2499866b3366498798ca95991c7f5267
    3 N2e458474c6ad455d9aa518d14c37c9ad
    4 N4c48c8df2ec4433cb16f85ef0ec9943c
    5 N6db4ae3b7dbd406fad973966fe2d43b6
    6 N7d3579216d3441ad870405cba5aa4eb2
    7 Naa27e62fe5c343b69ebac526ee56eb7f
    8 Nc326ef1856644575a67b7e2d3b82eb23
    9 Ndc6dceab408f44f5b0aae25068251d6b
    10 Ne4d9fab2a3924a76ab89abd45c11de4b
    11 anzsrc-for:01
    12 anzsrc-for:0104
    13 anzsrc-for:08
    14 anzsrc-for:0801
    15 schema:author N5e1cb663ef354384adf1df6d8440dd2f
    16 schema:citation sg:pub.10.1007/978-0-387-84858-7
    17 sg:pub.10.1007/978-1-4757-2440-0
    18 sg:pub.10.1007/978-3-540-35488-8_6
    19 sg:pub.10.1007/s00259-014-2961-x
    20 sg:pub.10.1007/s00259-017-3898-7
    21 sg:pub.10.1007/s00259-018-3977-4
    22 sg:pub.10.1007/s00259-018-4231-9
    23 sg:pub.10.1007/s00330-017-5302-1
    24 sg:pub.10.1023/a:1010933404324
    25 sg:pub.10.1038/nrclinonc.2017.141
    26 sg:pub.10.1038/s41598-017-19071-6
    27 sg:pub.10.1038/tpj.2010.57
    28 sg:pub.10.1186/1471-2105-7-91
    29 sg:pub.10.1186/1755-8794-5-23
    30 sg:pub.10.1186/s13040-017-0155-3
    31 schema:datePublished 2020-06-24
    32 schema:datePublishedReg 2020-06-24
    33 schema:description Multicenter studies are needed to demonstrate the clinical potential value of radiomics as a prognostic tool. However, variability in scanner models, acquisition protocols and reconstruction settings are unavoidable and radiomic features are notoriously sensitive to these factors, which hinders pooling them in a statistical analysis. A statistical harmonization method called ComBat was developed to deal with the “batch effect” in gene expression microarray data and was used in radiomics studies to deal with the “center-effect”. Our goal was to evaluate modifications in ComBat allowing for more flexibility in choosing a reference and improving robustness of the estimation. Two modified ComBat versions were evaluated: M-ComBat allows to transform all features distributions to a chosen reference, instead of the overall mean, providing more flexibility. B-ComBat adds bootstrap and Monte Carlo for improved robustness in the estimation. BM-ComBat combines both modifications. The four versions were compared regarding their ability to harmonize features in a multicenter context in two different clinical datasets. The first contains 119 locally advanced cervical cancer patients from 3 centers, with magnetic resonance imaging and positron emission tomography imaging. In that case ComBat was applied with 3 labels corresponding to each center. The second one contains 98 locally advanced laryngeal cancer patients from 5 centers with contrast-enhanced computed tomography. In that specific case, because imaging settings were highly heterogeneous even within each of the five centers, unsupervised clustering was used to determine two labels for applying ComBat. The impact of each harmonization was evaluated through three different machine learning pipelines for the modelling step in predicting the clinical outcomes, across two performance metrics (balanced accuracy and Matthews correlation coefficient). Before harmonization, almost all radiomic features had significantly different distributions between labels. These differences were successfully removed with all ComBat versions. The predictive ability of the radiomic models was always improved with harmonization and the improved ComBat provided the best results. This was observed consistently in both datasets, through all machine learning pipelines and performance metrics. The proposed modifications allow for more flexibility and robustness in the estimation. They also slightly but consistently improve the predictive power of resulting radiomic models.
    34 schema:genre article
    35 schema:isAccessibleForFree true
    36 schema:isPartOf Nbcfff7df04a5473eb7f8653972194485
    37 Nf7dad4f396e24c17b5802dfb08cfba49
    38 sg:journal.1045337
    39 schema:keywords Carlo
    40 Monte Carlo
    41 ability
    42 acquisition protocols
    43 advanced cervical cancer patients
    44 analysis
    45 batch effects
    46 better results
    47 bootstrap
    48 cancer patients
    49 cases
    50 center
    51 cervical cancer patients
    52 clinical datasets
    53 clinical outcomes
    54 clustering
    55 combat
    56 comparison
    57 computed tomography
    58 context
    59 contrast-enhanced computed tomography
    60 data
    61 dataset
    62 differences
    63 different clinical datasets
    64 different distributions
    65 different machine
    66 distribution
    67 effect
    68 emission tomography
    69 estimation
    70 expression microarray data
    71 factors
    72 feature distribution
    73 features
    74 flexibility
    75 gene expression microarray data
    76 goal
    77 harmonization
    78 harmonization methods
    79 imaging
    80 impact
    81 improved robustness
    82 labels
    83 laryngeal cancer patients
    84 machine
    85 magnetic resonance imaging
    86 means
    87 method
    88 metrics
    89 microarray data
    90 model
    91 modelling step
    92 modification
    93 more flexibility
    94 multicenter context
    95 multicenter study
    96 one
    97 outcomes
    98 overall mean
    99 patients
    100 performance comparison
    101 performance metrics
    102 pipeline
    103 positron emission tomography
    104 potential value
    105 power
    106 predictive ability
    107 predictive power
    108 prognostic tool
    109 protocol
    110 radiomic features
    111 radiomics
    112 radiomics model
    113 radiomics studies
    114 reconstruction settings
    115 reference
    116 resonance imaging
    117 results
    118 robustness
    119 scanner models
    120 second one
    121 setting
    122 specific case
    123 statistical analysis
    124 step
    125 study
    126 tomography
    127 tool
    128 unsupervised clustering
    129 values
    130 variability
    131 version
    132 schema:name Performance comparison of modified ComBat for harmonization of radiomic features for multicenter studies
    133 schema:pagination 10248
    134 schema:productId Ncab15ea428aa43c284334968ecc36042
    135 Nd464c976d9d1443b92eb24d47f478310
    136 Ne0038247c1724a7e9b0880cac25f47a1
    137 schema:sameAs https://app.dimensions.ai/details/publication/pub.1128717081
    138 https://doi.org/10.1038/s41598-020-66110-w
    139 schema:sdDatePublished 2022-08-04T17:10
    140 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    141 schema:sdPublisher Nde275bfa498c479ab2399dffda75c4c0
    142 schema:url https://doi.org/10.1038/s41598-020-66110-w
    143 sgo:license sg:explorer/license/
    144 sgo:sdDataset articles
    145 rdf:type schema:ScholarlyArticle
    146 N028e50742e1f48f3be36a3e5608a6772 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    147 schema:name Humans
    148 rdf:type schema:DefinedTerm
    149 N2499866b3366498798ca95991c7f5267 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    150 schema:name Female
    151 rdf:type schema:DefinedTerm
    152 N2e458474c6ad455d9aa518d14c37c9ad schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    153 schema:name Magnetic Resonance Imaging
    154 rdf:type schema:DefinedTerm
    155 N364de9811a3c4939bf3bc52cee5b37ed rdf:first sg:person.01206441206.87
    156 rdf:rest N59c24983cf734e05a418c9e537d52002
    157 N4c48c8df2ec4433cb16f85ef0ec9943c schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    158 schema:name Multicenter Studies as Topic
    159 rdf:type schema:DefinedTerm
    160 N4cf1e0bf9f834136b4bada2afb5da7f9 rdf:first sg:person.010361342027.13
    161 rdf:rest Nac25bd48af564384aba3b64ddda2f5dd
    162 N4dc085e9883644ccb962f8cbc5eb1853 rdf:first sg:person.0776411042.55
    163 rdf:rest Nbdff7cbff99643a5a8a3ee77e539471f
    164 N562c1119f9a34993880b8c0e11ac3000 rdf:first sg:person.01307140060.86
    165 rdf:rest Na6269368628f41738163789f771ae9d6
    166 N59c24983cf734e05a418c9e537d52002 rdf:first sg:person.01255045106.49
    167 rdf:rest Nfe4ea23fe31c40f8b7d4049beb22d83a
    168 N5e1cb663ef354384adf1df6d8440dd2f rdf:first sg:person.011560743623.06
    169 rdf:rest Nfec5206e42ec4f6f986ea70bce0681ad
    170 N63aec70d6c404e9798607d154edbcf2c rdf:first sg:person.07532416715.90
    171 rdf:rest Ndf738a9698f74d44b03a775fb41a00f2
    172 N6db4ae3b7dbd406fad973966fe2d43b6 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    173 schema:name Laryngeal Neoplasms
    174 rdf:type schema:DefinedTerm
    175 N7c31e6ee3a4a46d397fd222ce5b0fde8 rdf:first sg:person.01371121274.06
    176 rdf:rest N4cf1e0bf9f834136b4bada2afb5da7f9
    177 N7d3579216d3441ad870405cba5aa4eb2 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    178 schema:name Machine Learning
    179 rdf:type schema:DefinedTerm
    180 Na6269368628f41738163789f771ae9d6 rdf:first sg:person.01046015545.10
    181 rdf:rest Nfab0e68ebc774107a5fc34ae3228f6aa
    182 Na910a7765ae34aea843ea1aa3f2b37f0 rdf:first sg:person.0576143704.48
    183 rdf:rest N7c31e6ee3a4a46d397fd222ce5b0fde8
    184 Naa27e62fe5c343b69ebac526ee56eb7f schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    185 schema:name Positron-Emission Tomography
    186 rdf:type schema:DefinedTerm
    187 Nac25bd48af564384aba3b64ddda2f5dd rdf:first sg:person.01024223375.85
    188 rdf:rest N364de9811a3c4939bf3bc52cee5b37ed
    189 Nbcfff7df04a5473eb7f8653972194485 schema:volumeNumber 10
    190 rdf:type schema:PublicationVolume
    191 Nbdff7cbff99643a5a8a3ee77e539471f rdf:first sg:person.01134767703.78
    192 rdf:rest Na910a7765ae34aea843ea1aa3f2b37f0
    193 Nc326ef1856644575a67b7e2d3b82eb23 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    194 schema:name Radiographic Image Interpretation, Computer-Assisted
    195 rdf:type schema:DefinedTerm
    196 Ncab15ea428aa43c284334968ecc36042 schema:name doi
    197 schema:value 10.1038/s41598-020-66110-w
    198 rdf:type schema:PropertyValue
    199 Nd464c976d9d1443b92eb24d47f478310 schema:name pubmed_id
    200 schema:value 32581221
    201 rdf:type schema:PropertyValue
    202 Ndc6dceab408f44f5b0aae25068251d6b schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    203 schema:name Uterine Cervical Neoplasms
    204 rdf:type schema:DefinedTerm
    205 Nde275bfa498c479ab2399dffda75c4c0 schema:name Springer Nature - SN SciGraph project
    206 rdf:type schema:Organization
    207 Ndf738a9698f74d44b03a775fb41a00f2 rdf:first sg:person.0713341730.08
    208 rdf:rest N562c1119f9a34993880b8c0e11ac3000
    209 Ne0038247c1724a7e9b0880cac25f47a1 schema:name dimensions_id
    210 schema:value pub.1128717081
    211 rdf:type schema:PropertyValue
    212 Ne4d9fab2a3924a76ab89abd45c11de4b schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    213 schema:name Prognosis
    214 rdf:type schema:DefinedTerm
    215 Nf7dad4f396e24c17b5802dfb08cfba49 schema:issueNumber 1
    216 rdf:type schema:PublicationIssue
    217 Nfab0e68ebc774107a5fc34ae3228f6aa rdf:first sg:person.01240530667.47
    218 rdf:rest N4dc085e9883644ccb962f8cbc5eb1853
    219 Nfe4ea23fe31c40f8b7d4049beb22d83a rdf:first sg:person.01202724075.78
    220 rdf:rest rdf:nil
    221 Nfec5206e42ec4f6f986ea70bce0681ad rdf:first sg:person.01233533242.82
    222 rdf:rest N63aec70d6c404e9798607d154edbcf2c
    223 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
    224 schema:name Mathematical Sciences
    225 rdf:type schema:DefinedTerm
    226 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
    227 schema:name Statistics
    228 rdf:type schema:DefinedTerm
    229 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    230 schema:name Information and Computing Sciences
    231 rdf:type schema:DefinedTerm
    232 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
    233 schema:name Artificial Intelligence and Image Processing
    234 rdf:type schema:DefinedTerm
    235 sg:grant.7070033 http://pending.schema.org/fundedItem sg:pub.10.1038/s41598-020-66110-w
    236 rdf:type schema:MonetaryGrant
    237 sg:journal.1045337 schema:issn 2045-2322
    238 schema:name Scientific Reports
    239 schema:publisher Springer Nature
    240 rdf:type schema:Periodical
    241 sg:person.01024223375.85 schema:affiliation grid-institutes:grid.411766.3
    242 schema:familyName Pradier
    243 schema:givenName O.
    244 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01024223375.85
    245 rdf:type schema:Person
    246 sg:person.010361342027.13 schema:affiliation grid-institutes:None
    247 schema:familyName Rameé
    248 schema:givenName J. F.
    249 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010361342027.13
    250 rdf:type schema:Person
    251 sg:person.01046015545.10 schema:affiliation grid-institutes:grid.63984.30
    252 schema:familyName Alfieri
    253 schema:givenName J.
    254 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01046015545.10
    255 rdf:type schema:Person
    256 sg:person.01134767703.78 schema:affiliation grid-institutes:grid.63984.30
    257 schema:familyName Reinhold
    258 schema:givenName C.
    259 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01134767703.78
    260 rdf:type schema:Person
    261 sg:person.011560743623.06 schema:affiliation grid-institutes:grid.6289.5
    262 schema:familyName Da-ano
    263 schema:givenName R.
    264 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011560743623.06
    265 rdf:type schema:Person
    266 sg:person.01202724075.78 schema:affiliation grid-institutes:grid.6289.5
    267 schema:familyName Hatt
    268 schema:givenName M.
    269 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01202724075.78
    270 rdf:type schema:Person
    271 sg:person.01206441206.87 schema:affiliation grid-institutes:grid.411766.3
    272 schema:familyName Schick
    273 schema:givenName U.
    274 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01206441206.87
    275 rdf:type schema:Person
    276 sg:person.01233533242.82 schema:affiliation grid-institutes:None
    277 schema:familyName Masson
    278 schema:givenName I.
    279 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01233533242.82
    280 rdf:type schema:Person
    281 sg:person.01240530667.47 schema:affiliation grid-institutes:grid.4817.a
    282 schema:familyName Rousseau
    283 schema:givenName C.
    284 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01240530667.47
    285 rdf:type schema:Person
    286 sg:person.01255045106.49 schema:affiliation grid-institutes:grid.6289.5
    287 schema:familyName Visvikis
    288 schema:givenName D.
    289 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01255045106.49
    290 rdf:type schema:Person
    291 sg:person.01307140060.86 schema:affiliation grid-institutes:grid.6289.5
    292 schema:familyName Robin
    293 schema:givenName P.
    294 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01307140060.86
    295 rdf:type schema:Person
    296 sg:person.01371121274.06 schema:affiliation grid-institutes:grid.463996.7
    297 schema:familyName De Crevoisier
    298 schema:givenName R.
    299 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01371121274.06
    300 rdf:type schema:Person
    301 sg:person.0576143704.48 schema:affiliation grid-institutes:grid.463996.7
    302 schema:familyName Castelli
    303 schema:givenName J.
    304 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0576143704.48
    305 rdf:type schema:Person
    306 sg:person.0713341730.08 schema:affiliation grid-institutes:None
    307 schema:familyName Doré
    308 schema:givenName M.
    309 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0713341730.08
    310 rdf:type schema:Person
    311 sg:person.07532416715.90 schema:affiliation grid-institutes:grid.411766.3
    312 schema:familyName Lucia
    313 schema:givenName F.
    314 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07532416715.90
    315 rdf:type schema:Person
    316 sg:person.0776411042.55 schema:affiliation grid-institutes:None
    317 schema:familyName Mervoyer
    318 schema:givenName A.
    319 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0776411042.55
    320 rdf:type schema:Person
    321 sg:pub.10.1007/978-0-387-84858-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1032573094
    322 https://doi.org/10.1007/978-0-387-84858-7
    323 rdf:type schema:CreativeWork
    324 sg:pub.10.1007/978-1-4757-2440-0 schema:sameAs https://app.dimensions.ai/details/publication/pub.1027312764
    325 https://doi.org/10.1007/978-1-4757-2440-0
    326 rdf:type schema:CreativeWork
    327 sg:pub.10.1007/978-3-540-35488-8_6 schema:sameAs https://app.dimensions.ai/details/publication/pub.1031342354
    328 https://doi.org/10.1007/978-3-540-35488-8_6
    329 rdf:type schema:CreativeWork
    330 sg:pub.10.1007/s00259-014-2961-x schema:sameAs https://app.dimensions.ai/details/publication/pub.1008918527
    331 https://doi.org/10.1007/s00259-014-2961-x
    332 rdf:type schema:CreativeWork
    333 sg:pub.10.1007/s00259-017-3898-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1099696659
    334 https://doi.org/10.1007/s00259-017-3898-7
    335 rdf:type schema:CreativeWork
    336 sg:pub.10.1007/s00259-018-3977-4 schema:sameAs https://app.dimensions.ai/details/publication/pub.1101318626
    337 https://doi.org/10.1007/s00259-018-3977-4
    338 rdf:type schema:CreativeWork
    339 sg:pub.10.1007/s00259-018-4231-9 schema:sameAs https://app.dimensions.ai/details/publication/pub.1110448603
    340 https://doi.org/10.1007/s00259-018-4231-9
    341 rdf:type schema:CreativeWork
    342 sg:pub.10.1007/s00330-017-5302-1 schema:sameAs https://app.dimensions.ai/details/publication/pub.1101678461
    343 https://doi.org/10.1007/s00330-017-5302-1
    344 rdf:type schema:CreativeWork
    345 sg:pub.10.1023/a:1010933404324 schema:sameAs https://app.dimensions.ai/details/publication/pub.1024739340
    346 https://doi.org/10.1023/a:1010933404324
    347 rdf:type schema:CreativeWork
    348 sg:pub.10.1038/nrclinonc.2017.141 schema:sameAs https://app.dimensions.ai/details/publication/pub.1092061102
    349 https://doi.org/10.1038/nrclinonc.2017.141
    350 rdf:type schema:CreativeWork
    351 sg:pub.10.1038/s41598-017-19071-6 schema:sameAs https://app.dimensions.ai/details/publication/pub.1100212558
    352 https://doi.org/10.1038/s41598-017-19071-6
    353 rdf:type schema:CreativeWork
    354 sg:pub.10.1038/tpj.2010.57 schema:sameAs https://app.dimensions.ai/details/publication/pub.1030763566
    355 https://doi.org/10.1038/tpj.2010.57
    356 rdf:type schema:CreativeWork
    357 sg:pub.10.1186/1471-2105-7-91 schema:sameAs https://app.dimensions.ai/details/publication/pub.1034791610
    358 https://doi.org/10.1186/1471-2105-7-91
    359 rdf:type schema:CreativeWork
    360 sg:pub.10.1186/1755-8794-5-23 schema:sameAs https://app.dimensions.ai/details/publication/pub.1016278320
    361 https://doi.org/10.1186/1755-8794-5-23
    362 rdf:type schema:CreativeWork
    363 sg:pub.10.1186/s13040-017-0155-3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1099596880
    364 https://doi.org/10.1186/s13040-017-0155-3
    365 rdf:type schema:CreativeWork
    366 grid-institutes:None schema:alternateName Department of Medical Oncology, Centre Hospitalier de Vendee, La Roche sur Yon, France
    367 Department of Radiation Oncology, Institut de cancérologie de l’Ouest René-Gauducheau, Saint-Herblain, France
    368 schema:name Department of Medical Oncology, Centre Hospitalier de Vendee, La Roche sur Yon, France
    369 Department of Radiation Oncology, Institut de cancérologie de l’Ouest René-Gauducheau, Saint-Herblain, France
    370 INSERM, UMR 1101, LaTIM, University of Brest, Brest, France
    371 rdf:type schema:Organization
    372 grid-institutes:grid.411766.3 schema:alternateName Radiation Oncology Department, University Hospital, Brest, France
    373 schema:name INSERM, UMR 1101, LaTIM, University of Brest, Brest, France
    374 Radiation Oncology Department, University Hospital, Brest, France
    375 rdf:type schema:Organization
    376 grid-institutes:grid.463996.7 schema:alternateName University of Rennes 1, LTSI, Rennes, France
    377 schema:name Radiotherapy Department Cancer, Institute Eugene Marquis, Rennes, France
    378 University of Rennes 1, LTSI, Rennes, France
    379 rdf:type schema:Organization
    380 grid-institutes:grid.4817.a schema:alternateName CRCINA, University of Nantes, INSERM UMR1232, CNRS-ERL6001, Nantes, France
    381 schema:name CRCINA, University of Nantes, INSERM UMR1232, CNRS-ERL6001, Nantes, France
    382 Department of Nuclear Medicine, Institut de cancerologie de l’Ouest René-Gauducheau, Saint-Herblain, France
    383 rdf:type schema:Organization
    384 grid-institutes:grid.6289.5 schema:alternateName Department of Nuclear Medicine, University of Brest, Brest, France
    385 INSERM, UMR 1101, LaTIM, University of Brest, Brest, France
    386 schema:name Department of Nuclear Medicine, University of Brest, Brest, France
    387 INSERM, UMR 1101, LaTIM, University of Brest, Brest, France
    388 rdf:type schema:Organization
    389 grid-institutes:grid.63984.30 schema:alternateName Department of Radiation Oncology, McGill University Health Centre, Montreal, Quebec, Canada
    390 Department of Radiology, McGill University Health Centre, Montreal, Canada
    391 schema:name Department of Radiation Oncology, McGill University Health Centre, Montreal, Quebec, Canada
    392 Department of Radiology, McGill University Health Centre, Montreal, Canada
    393 rdf:type schema:Organization
     




    Preview window. Press ESC to close (or click here)


    ...