Machine learning automatically detects COVID-19 using chest CTs in a large multicenter cohort View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2021-05-01

AUTHORS

Eduardo J. Mortani Barbosa, Bogdan Georgescu, Shikha Chaganti, Gorka Bastarrika Aleman, Jordi Broncano Cabrero, Guillaume Chabin, Thomas Flohr, Philippe Grenier, Sasa Grbic, Nakul Gupta, François Mellot, Savvas Nicolaou, Thomas Re, Pina Sanelli, Alexander W. Sauter, Youngjin Yoo, Valentin Ziebandt, Dorin Comaniciu

ABSTRACT

OBJECTIVES: To investigate machine learning classifiers and interpretable models using chest CT for detection of COVID-19 and differentiation from other pneumonias, interstitial lung disease (ILD) and normal CTs. METHODS: Our retrospective multi-institutional study obtained 2446 chest CTs from 16 institutions (including 1161 COVID-19 patients). Training/validation/testing cohorts included 1011/50/100 COVID-19, 388/16/33 ILD, 189/16/33 other pneumonias, and 559/17/34 normal (no pathologies) CTs. A metric-based approach for the classification of COVID-19 used interpretable features, relying on logistic regression and random forests. A deep learning-based classifier differentiated COVID-19 via 3D features extracted directly from CT attenuation and probability distribution of airspace opacities. RESULTS: Most discriminative features of COVID-19 are the percentage of airspace opacity and peripheral and basal predominant opacities, concordant with the typical characterization of COVID-19 in the literature. Unsupervised hierarchical clustering compares feature distribution across COVID-19 and control cohorts. The metrics-based classifier achieved AUC = 0.83, sensitivity = 0.74, and specificity = 0.79 versus respectively 0.93, 0.90, and 0.83 for the DL-based classifier. Most of ambiguity comes from non-COVID-19 pneumonia with manifestations that overlap with COVID-19, as well as mild COVID-19 cases. Non-COVID-19 classification performance is 91% for ILD, 64% for other pneumonias, and 94% for no pathologies, which demonstrates the robustness of our method against different compositions of control groups. CONCLUSIONS: Our new method accurately discriminates COVID-19 from other types of pneumonia, ILD, and CTs with no pathologies, using quantitative imaging features derived from chest CT, while balancing interpretability of results and classification performance and, therefore, may be useful to facilitate diagnosis of COVID-19. KEY POINTS: • Unsupervised clustering reveals the key tomographic features including percent airspace opacity and peripheral and basal opacities most typical of COVID-19 relative to control groups. • COVID-19-positive CTs were compared with COVID-19-negative chest CTs (including a balanced distribution of non-COVID-19 pneumonia, ILD, and no pathologies). Classification accuracies for COVID-19, pneumonia, ILD, and CT scans with no pathologies are respectively 90%, 64%, 91%, and 94%. • Our deep learning (DL)-based classification method demonstrates an AUC of 0.93 (sensitivity 90%, specificity 83%). Machine learning methods applied to quantitative chest CT metrics can therefore improve diagnostic accuracy in suspected COVID-19, particularly in resource-constrained environments. More... »

PAGES

1-11

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s00330-021-07937-3

DOI

http://dx.doi.org/10.1007/s00330-021-07937-3

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1137686426

PUBMED

https://www.ncbi.nlm.nih.gov/pubmed/33934177


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/2008/11", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Medical and Health Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/1103", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Clinical Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "COVID-19", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Humans", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Machine Learning", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Retrospective Studies", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "SARS-CoV-2", 
        "type": "DefinedTerm"
      }, 
      {
        "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
        "name": "Thorax", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Penn Medicine, Philadelphia, PA USA", 
          "id": "http://www.grid.ac/institutes/grid.412701.1", 
          "name": [
            "Division of Cardiothoracic Imaging, Department of Radiology, Perelman School of Medicine, University of Pennsylvania, 3400 Spruce Street, Ground Floor Founders Bldg, Philadelphia, PA 19104 USA", 
            "Penn Medicine, Philadelphia, PA USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Barbosa", 
        "givenName": "Eduardo J. Mortani", 
        "id": "sg:person.01320676335.36", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01320676335.36"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Princeton, NJ USA", 
          "id": "http://www.grid.ac/institutes/grid.415886.6", 
          "name": [
            "Siemens Healthineers, Princeton, NJ USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Georgescu", 
        "givenName": "Bogdan", 
        "id": "sg:person.0703547214.37", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0703547214.37"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Princeton, NJ USA", 
          "id": "http://www.grid.ac/institutes/grid.415886.6", 
          "name": [
            "Siemens Healthineers, Princeton, NJ USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Chaganti", 
        "givenName": "Shikha", 
        "id": "sg:person.016574365001.55", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016574365001.55"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Cl\u00ednica Universidad de Navarra, Navarra, Spain", 
          "id": "http://www.grid.ac/institutes/grid.411730.0", 
          "name": [
            "Cl\u00ednica Universidad de Navarra, Navarra, Spain"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Aleman", 
        "givenName": "Gorka Bastarrika", 
        "id": "sg:person.016306171101.99", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016306171101.99"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Health Time, Ja\u00e9n, Spain", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Health Time, Ja\u00e9n, Spain"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Cabrero", 
        "givenName": "Jordi Broncano", 
        "id": "sg:person.010652722323.96", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010652722323.96"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Paris, France", 
          "id": "http://www.grid.ac/institutes/None", 
          "name": [
            "Siemens Healthineers, Paris, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Chabin", 
        "givenName": "Guillaume", 
        "id": "sg:person.013722107237.15", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013722107237.15"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Forchheim, Germany", 
          "id": "http://www.grid.ac/institutes/grid.481749.7", 
          "name": [
            "Siemens Healthineers, Forchheim, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Flohr", 
        "givenName": "Thomas", 
        "id": "sg:person.0754131024.61", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0754131024.61"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "H\u00f4pital Foch, Suresnes, France", 
          "id": "http://www.grid.ac/institutes/grid.414106.6", 
          "name": [
            "H\u00f4pital Foch, Suresnes, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Grenier", 
        "givenName": "Philippe", 
        "id": "sg:person.01050753505.86", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01050753505.86"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Princeton, NJ USA", 
          "id": "http://www.grid.ac/institutes/grid.415886.6", 
          "name": [
            "Siemens Healthineers, Princeton, NJ USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Grbic", 
        "givenName": "Sasa", 
        "id": "sg:person.01126227511.07", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01126227511.07"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Houston Methodist, Houston, TX USA", 
          "id": "http://www.grid.ac/institutes/grid.63368.38", 
          "name": [
            "Houston Methodist, Houston, TX USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Gupta", 
        "givenName": "Nakul", 
        "id": "sg:person.011402347273.82", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011402347273.82"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "H\u00f4pital Foch, Suresnes, France", 
          "id": "http://www.grid.ac/institutes/grid.414106.6", 
          "name": [
            "H\u00f4pital Foch, Suresnes, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Mellot", 
        "givenName": "Fran\u00e7ois", 
        "id": "sg:person.01366452774.22", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01366452774.22"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Vancouver General Hospital, Vancouver, BC Canada", 
          "id": "http://www.grid.ac/institutes/grid.412541.7", 
          "name": [
            "Vancouver General Hospital, Vancouver, BC Canada"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Nicolaou", 
        "givenName": "Savvas", 
        "id": "sg:person.01225416107.94", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01225416107.94"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Princeton, NJ USA", 
          "id": "http://www.grid.ac/institutes/grid.415886.6", 
          "name": [
            "Siemens Healthineers, Princeton, NJ USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Re", 
        "givenName": "Thomas", 
        "id": "sg:person.012162673120.78", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012162673120.78"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Donald and Barbara Zucker School of Medicine, Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY USA", 
          "id": "http://www.grid.ac/institutes/grid.416477.7", 
          "name": [
            "Donald and Barbara Zucker School of Medicine, Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Sanelli", 
        "givenName": "Pina", 
        "id": "sg:person.0600344727.20", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0600344727.20"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University Hospital Basel, Clinic of Radiology & Nuclear medicine, Basel, Switzerland", 
          "id": "http://www.grid.ac/institutes/grid.410567.1", 
          "name": [
            "University Hospital Basel, Clinic of Radiology & Nuclear medicine, Basel, Switzerland"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Sauter", 
        "givenName": "Alexander W.", 
        "id": "sg:person.0655016751.47", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0655016751.47"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Princeton, NJ USA", 
          "id": "http://www.grid.ac/institutes/grid.415886.6", 
          "name": [
            "Siemens Healthineers, Princeton, NJ USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Yoo", 
        "givenName": "Youngjin", 
        "id": "sg:person.011735325401.24", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011735325401.24"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Forchheim, Germany", 
          "id": "http://www.grid.ac/institutes/grid.481749.7", 
          "name": [
            "Siemens Healthineers, Forchheim, Germany"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Ziebandt", 
        "givenName": "Valentin", 
        "id": "sg:person.012532706001.83", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012532706001.83"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Siemens Healthineers, Princeton, NJ USA", 
          "id": "http://www.grid.ac/institutes/grid.415886.6", 
          "name": [
            "Siemens Healthineers, Princeton, NJ USA"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Comaniciu", 
        "givenName": "Dorin", 
        "id": "sg:person.01066111014.77", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01066111014.77"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/s00330-020-07273-y", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1131095776", 
          "https://doi.org/10.1007/s00330-020-07273-y"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1038/s41467-020-17971-2", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1130092582", 
          "https://doi.org/10.1038/s41467-020-17971-2"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s10096-020-03901-z", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1127150720", 
          "https://doi.org/10.1007/s10096-020-03901-z"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2021-05-01", 
    "datePublishedReg": "2021-05-01", 
    "description": "OBJECTIVES: To investigate machine learning classifiers and interpretable models using chest CT for detection of COVID-19 and differentiation from other pneumonias, interstitial lung disease (ILD) and normal CTs.\nMETHODS: Our retrospective multi-institutional study obtained 2446 chest CTs from 16 institutions (including 1161 COVID-19 patients). Training/validation/testing cohorts included 1011/50/100 COVID-19, 388/16/33 ILD, 189/16/33 other pneumonias, and 559/17/34 normal (no pathologies) CTs. A metric-based approach for the classification of COVID-19 used interpretable features, relying on logistic regression and random forests. A deep learning-based classifier differentiated COVID-19 via 3D features extracted directly from CT attenuation and probability distribution of airspace opacities.\nRESULTS: Most discriminative features of COVID-19 are the percentage of airspace opacity and peripheral and basal predominant opacities, concordant with the typical characterization of COVID-19 in the literature. Unsupervised hierarchical clustering compares feature distribution across COVID-19 and control cohorts. The metrics-based classifier achieved AUC = 0.83, sensitivity = 0.74, and specificity = 0.79 versus respectively 0.93, 0.90, and 0.83 for the DL-based classifier. Most of ambiguity comes from non-COVID-19 pneumonia with manifestations that overlap with COVID-19, as well as mild COVID-19 cases. Non-COVID-19 classification performance is 91% for ILD, 64% for other pneumonias, and 94% for no pathologies, which demonstrates the robustness of our method against different compositions of control groups.\nCONCLUSIONS: Our new method accurately discriminates COVID-19 from other types of pneumonia, ILD, and CTs with no pathologies, using quantitative imaging features derived from chest CT, while balancing interpretability of results and classification performance and, therefore, may be useful to facilitate diagnosis of COVID-19.\nKEY POINTS: \u2022 Unsupervised clustering reveals the key tomographic features including percent airspace opacity and peripheral and basal opacities most typical of COVID-19 relative to control groups. \u2022 COVID-19-positive CTs were compared with COVID-19-negative chest CTs (including a balanced distribution of non-COVID-19 pneumonia, ILD, and no pathologies). Classification accuracies for COVID-19, pneumonia, ILD, and CT scans with no pathologies are respectively 90%, 64%, 91%, and 94%. \u2022 Our deep learning (DL)-based classification method demonstrates an AUC of 0.93 (sensitivity 90%, specificity 83%). Machine learning methods applied to quantitative chest CT metrics can therefore improve diagnostic accuracy in suspected COVID-19, particularly in resource-constrained environments.", 
    "genre": "article", 
    "id": "sg:pub.10.1007/s00330-021-07937-3", 
    "inLanguage": "en", 
    "isAccessibleForFree": true, 
    "isPartOf": [
      {
        "id": "sg:journal.1289120", 
        "issn": [
          "0938-7994", 
          "1432-1084"
        ], 
        "name": "European Radiology", 
        "publisher": "Springer Nature", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "11", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "31"
      }
    ], 
    "keywords": [
      "interstitial lung disease", 
      "chest CT", 
      "airspace opacities", 
      "normal CT", 
      "COVID-19", 
      "mild COVID-19 cases", 
      "retrospective multi-institutional study", 
      "large multicenter cohort", 
      "type of pneumonia", 
      "multi-institutional study", 
      "COVID-19 relative", 
      "quantitative imaging features", 
      "COVID-19 cases", 
      "multicenter cohort", 
      "lung disease", 
      "control cohort", 
      "imaging features", 
      "tomographic features", 
      "CT metrics", 
      "Unsupervised hierarchical clustering", 
      "pneumonia", 
      "testing cohort", 
      "control group", 
      "logistic regression", 
      "CT attenuation", 
      "diagnostic accuracy", 
      "CT", 
      "cohort", 
      "pathology", 
      "AUC", 
      "opacity", 
      "group", 
      "disease", 
      "dl", 
      "diagnosis", 
      "manifestations", 
      "regression", 
      "specificity", 
      "percentage", 
      "differentiation", 
      "relatives", 
      "features", 
      "resource-constrained environments", 
      "study", 
      "cases", 
      "interpretability of results", 
      "sensitivity", 
      "unsupervised clustering", 
      "hierarchical clustering", 
      "attenuation", 
      "institutions", 
      "method", 
      "literature", 
      "detection", 
      "classification", 
      "types", 
      "typical characterization", 
      "deep learning-based classifiers", 
      "results", 
      "learning-based classifiers", 
      "distribution", 
      "accuracy", 
      "model", 
      "approach", 
      "characterization", 
      "clustering", 
      "random forest", 
      "machine learning methods", 
      "new method", 
      "classification performance", 
      "composition", 
      "interpretability", 
      "classification accuracy", 
      "performance", 
      "learning", 
      "classifier", 
      "environment", 
      "interpretable features", 
      "interpretable models", 
      "metrics", 
      "classification method", 
      "deep learning", 
      "machine", 
      "different compositions", 
      "ambiguity", 
      "learning method", 
      "robustness", 
      "metric-based approach", 
      "feature distributions", 
      "probability distribution", 
      "forest", 
      "Training/validation/testing cohorts", 
      "validation/testing cohorts", 
      "basal predominant opacities", 
      "predominant opacities", 
      "metrics-based classifier", 
      "key tomographic features", 
      "percent airspace opacity", 
      "basal opacities", 
      "COVID-19-positive CTs", 
      "COVID-19-negative chest CTs", 
      "quantitative chest CT metrics", 
      "chest CT metrics"
    ], 
    "name": "Machine learning automatically detects COVID-19 using chest CTs in a large multicenter cohort", 
    "pagination": "1-11", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1137686426"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s00330-021-07937-3"
        ]
      }, 
      {
        "name": "pubmed_id", 
        "type": "PropertyValue", 
        "value": [
          "33934177"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s00330-021-07937-3", 
      "https://app.dimensions.ai/details/publication/pub.1137686426"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2022-01-01T19:03", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20220101/entities/gbq_results/article/article_918.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://doi.org/10.1007/s00330-021-07937-3"
  }
]
 

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.10.1007/s00330-021-07937-3'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/s00330-021-07937-3'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s00330-021-07937-3'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s00330-021-07937-3'


 

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

350 TRIPLES      22 PREDICATES      138 URIs      127 LITERALS      13 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s00330-021-07937-3 schema:about N0c69db6065c54d95b3df0ae628862ae3
2 N70571038bcfc4e1fa07f5f5d5c4eafa3
3 N7fe52ce529a54b438ee16be941177915
4 N8b05abcfae5f4104aee13873dae8224a
5 Ne5cb7dd872c4413eb013a0c9fbcc184c
6 Ne6356da21eed48e28c06dd9e6be940a4
7 anzsrc-for:11
8 anzsrc-for:1103
9 schema:author N16df7a20cc224f40b7bf0bff7b894a83
10 schema:citation sg:pub.10.1007/s00330-020-07273-y
11 sg:pub.10.1007/s10096-020-03901-z
12 sg:pub.10.1038/s41467-020-17971-2
13 schema:datePublished 2021-05-01
14 schema:datePublishedReg 2021-05-01
15 schema:description OBJECTIVES: To investigate machine learning classifiers and interpretable models using chest CT for detection of COVID-19 and differentiation from other pneumonias, interstitial lung disease (ILD) and normal CTs. METHODS: Our retrospective multi-institutional study obtained 2446 chest CTs from 16 institutions (including 1161 COVID-19 patients). Training/validation/testing cohorts included 1011/50/100 COVID-19, 388/16/33 ILD, 189/16/33 other pneumonias, and 559/17/34 normal (no pathologies) CTs. A metric-based approach for the classification of COVID-19 used interpretable features, relying on logistic regression and random forests. A deep learning-based classifier differentiated COVID-19 via 3D features extracted directly from CT attenuation and probability distribution of airspace opacities. RESULTS: Most discriminative features of COVID-19 are the percentage of airspace opacity and peripheral and basal predominant opacities, concordant with the typical characterization of COVID-19 in the literature. Unsupervised hierarchical clustering compares feature distribution across COVID-19 and control cohorts. The metrics-based classifier achieved AUC = 0.83, sensitivity = 0.74, and specificity = 0.79 versus respectively 0.93, 0.90, and 0.83 for the DL-based classifier. Most of ambiguity comes from non-COVID-19 pneumonia with manifestations that overlap with COVID-19, as well as mild COVID-19 cases. Non-COVID-19 classification performance is 91% for ILD, 64% for other pneumonias, and 94% for no pathologies, which demonstrates the robustness of our method against different compositions of control groups. CONCLUSIONS: Our new method accurately discriminates COVID-19 from other types of pneumonia, ILD, and CTs with no pathologies, using quantitative imaging features derived from chest CT, while balancing interpretability of results and classification performance and, therefore, may be useful to facilitate diagnosis of COVID-19. KEY POINTS: • Unsupervised clustering reveals the key tomographic features including percent airspace opacity and peripheral and basal opacities most typical of COVID-19 relative to control groups. • COVID-19-positive CTs were compared with COVID-19-negative chest CTs (including a balanced distribution of non-COVID-19 pneumonia, ILD, and no pathologies). Classification accuracies for COVID-19, pneumonia, ILD, and CT scans with no pathologies are respectively 90%, 64%, 91%, and 94%. • Our deep learning (DL)-based classification method demonstrates an AUC of 0.93 (sensitivity 90%, specificity 83%). Machine learning methods applied to quantitative chest CT metrics can therefore improve diagnostic accuracy in suspected COVID-19, particularly in resource-constrained environments.
16 schema:genre article
17 schema:inLanguage en
18 schema:isAccessibleForFree true
19 schema:isPartOf N30ed9c801c774a5cbd27dae2707623eb
20 N413e1217cfe449a3bc01a4122d9329dc
21 sg:journal.1289120
22 schema:keywords AUC
23 COVID-19
24 COVID-19 cases
25 COVID-19 relative
26 COVID-19-negative chest CTs
27 COVID-19-positive CTs
28 CT
29 CT attenuation
30 CT metrics
31 Training/validation/testing cohorts
32 Unsupervised hierarchical clustering
33 accuracy
34 airspace opacities
35 ambiguity
36 approach
37 attenuation
38 basal opacities
39 basal predominant opacities
40 cases
41 characterization
42 chest CT
43 chest CT metrics
44 classification
45 classification accuracy
46 classification method
47 classification performance
48 classifier
49 clustering
50 cohort
51 composition
52 control cohort
53 control group
54 deep learning
55 deep learning-based classifiers
56 detection
57 diagnosis
58 diagnostic accuracy
59 different compositions
60 differentiation
61 disease
62 distribution
63 dl
64 environment
65 feature distributions
66 features
67 forest
68 group
69 hierarchical clustering
70 imaging features
71 institutions
72 interpretability
73 interpretability of results
74 interpretable features
75 interpretable models
76 interstitial lung disease
77 key tomographic features
78 large multicenter cohort
79 learning
80 learning method
81 learning-based classifiers
82 literature
83 logistic regression
84 lung disease
85 machine
86 machine learning methods
87 manifestations
88 method
89 metric-based approach
90 metrics
91 metrics-based classifier
92 mild COVID-19 cases
93 model
94 multi-institutional study
95 multicenter cohort
96 new method
97 normal CT
98 opacity
99 pathology
100 percent airspace opacity
101 percentage
102 performance
103 pneumonia
104 predominant opacities
105 probability distribution
106 quantitative chest CT metrics
107 quantitative imaging features
108 random forest
109 regression
110 relatives
111 resource-constrained environments
112 results
113 retrospective multi-institutional study
114 robustness
115 sensitivity
116 specificity
117 study
118 testing cohort
119 tomographic features
120 type of pneumonia
121 types
122 typical characterization
123 unsupervised clustering
124 validation/testing cohorts
125 schema:name Machine learning automatically detects COVID-19 using chest CTs in a large multicenter cohort
126 schema:pagination 1-11
127 schema:productId N4001cdea5f6e441c857ca030cf347d51
128 N900016c5b0784f9d8ed74d69e17d9023
129 Ne92ea760f64846429c6d0ee6db47a841
130 schema:sameAs https://app.dimensions.ai/details/publication/pub.1137686426
131 https://doi.org/10.1007/s00330-021-07937-3
132 schema:sdDatePublished 2022-01-01T19:03
133 schema:sdLicense https://scigraph.springernature.com/explorer/license/
134 schema:sdPublisher Nbf330f37917e4762aaf7baab0b2b752c
135 schema:url https://doi.org/10.1007/s00330-021-07937-3
136 sgo:license sg:explorer/license/
137 sgo:sdDataset articles
138 rdf:type schema:ScholarlyArticle
139 N023b978b91a64186b9b2e060c766a88a rdf:first sg:person.0600344727.20
140 rdf:rest Nea9fc583187a41e587e76aa5267b455b
141 N02895f23dc534b2e9181c34247cf403f rdf:first sg:person.01366452774.22
142 rdf:rest Nb94c4b2de1734f83930233ecd8571b76
143 N0aab0351c7e14ebdb5299dd11599df1c rdf:first sg:person.011402347273.82
144 rdf:rest N02895f23dc534b2e9181c34247cf403f
145 N0af3f77cf8e14099af1d57eeb9ea817c rdf:first sg:person.016574365001.55
146 rdf:rest N577004167ddd482ea72b8d912ab4219c
147 N0c69db6065c54d95b3df0ae628862ae3 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
148 schema:name Machine Learning
149 rdf:type schema:DefinedTerm
150 N106d1182096849ab828baf701c2058f0 rdf:first sg:person.01066111014.77
151 rdf:rest rdf:nil
152 N16df7a20cc224f40b7bf0bff7b894a83 rdf:first sg:person.01320676335.36
153 rdf:rest Nfef9863c4539454b8844d4fa8aa9393e
154 N30ed9c801c774a5cbd27dae2707623eb schema:issueNumber 11
155 rdf:type schema:PublicationIssue
156 N4001cdea5f6e441c857ca030cf347d51 schema:name dimensions_id
157 schema:value pub.1137686426
158 rdf:type schema:PropertyValue
159 N413e1217cfe449a3bc01a4122d9329dc schema:volumeNumber 31
160 rdf:type schema:PublicationVolume
161 N4afdd615f05947d89cf254e6e533b305 rdf:first sg:person.011735325401.24
162 rdf:rest Nddb8496a6fee4b9e83792fa37454fdbe
163 N577004167ddd482ea72b8d912ab4219c rdf:first sg:person.016306171101.99
164 rdf:rest Nb198f129f043415ba939a8918c831de7
165 N6231332d53124b9b9a22c13b4e9aa9b5 rdf:first sg:person.0754131024.61
166 rdf:rest Nfefb0a9eed9c46299151f34422351b00
167 N70571038bcfc4e1fa07f5f5d5c4eafa3 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
168 schema:name Humans
169 rdf:type schema:DefinedTerm
170 N7fe52ce529a54b438ee16be941177915 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
171 schema:name Thorax
172 rdf:type schema:DefinedTerm
173 N86af7b36d57b4670b7d4ce4f0dada56c rdf:first sg:person.013722107237.15
174 rdf:rest N6231332d53124b9b9a22c13b4e9aa9b5
175 N8b05abcfae5f4104aee13873dae8224a schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
176 schema:name Retrospective Studies
177 rdf:type schema:DefinedTerm
178 N900016c5b0784f9d8ed74d69e17d9023 schema:name pubmed_id
179 schema:value 33934177
180 rdf:type schema:PropertyValue
181 N9f9539133c13485a93863eca5e7b8c5f rdf:first sg:person.01126227511.07
182 rdf:rest N0aab0351c7e14ebdb5299dd11599df1c
183 Na8b1a0dc20454206bd9072c363621fd7 rdf:first sg:person.012162673120.78
184 rdf:rest N023b978b91a64186b9b2e060c766a88a
185 Nb198f129f043415ba939a8918c831de7 rdf:first sg:person.010652722323.96
186 rdf:rest N86af7b36d57b4670b7d4ce4f0dada56c
187 Nb94c4b2de1734f83930233ecd8571b76 rdf:first sg:person.01225416107.94
188 rdf:rest Na8b1a0dc20454206bd9072c363621fd7
189 Nbf330f37917e4762aaf7baab0b2b752c schema:name Springer Nature - SN SciGraph project
190 rdf:type schema:Organization
191 Nddb8496a6fee4b9e83792fa37454fdbe rdf:first sg:person.012532706001.83
192 rdf:rest N106d1182096849ab828baf701c2058f0
193 Ne5cb7dd872c4413eb013a0c9fbcc184c schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
194 schema:name COVID-19
195 rdf:type schema:DefinedTerm
196 Ne6356da21eed48e28c06dd9e6be940a4 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
197 schema:name SARS-CoV-2
198 rdf:type schema:DefinedTerm
199 Ne92ea760f64846429c6d0ee6db47a841 schema:name doi
200 schema:value 10.1007/s00330-021-07937-3
201 rdf:type schema:PropertyValue
202 Nea9fc583187a41e587e76aa5267b455b rdf:first sg:person.0655016751.47
203 rdf:rest N4afdd615f05947d89cf254e6e533b305
204 Nfef9863c4539454b8844d4fa8aa9393e rdf:first sg:person.0703547214.37
205 rdf:rest N0af3f77cf8e14099af1d57eeb9ea817c
206 Nfefb0a9eed9c46299151f34422351b00 rdf:first sg:person.01050753505.86
207 rdf:rest N9f9539133c13485a93863eca5e7b8c5f
208 anzsrc-for:11 schema:inDefinedTermSet anzsrc-for:
209 schema:name Medical and Health Sciences
210 rdf:type schema:DefinedTerm
211 anzsrc-for:1103 schema:inDefinedTermSet anzsrc-for:
212 schema:name Clinical Sciences
213 rdf:type schema:DefinedTerm
214 sg:journal.1289120 schema:issn 0938-7994
215 1432-1084
216 schema:name European Radiology
217 schema:publisher Springer Nature
218 rdf:type schema:Periodical
219 sg:person.01050753505.86 schema:affiliation grid-institutes:grid.414106.6
220 schema:familyName Grenier
221 schema:givenName Philippe
222 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01050753505.86
223 rdf:type schema:Person
224 sg:person.010652722323.96 schema:affiliation grid-institutes:None
225 schema:familyName Cabrero
226 schema:givenName Jordi Broncano
227 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010652722323.96
228 rdf:type schema:Person
229 sg:person.01066111014.77 schema:affiliation grid-institutes:grid.415886.6
230 schema:familyName Comaniciu
231 schema:givenName Dorin
232 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01066111014.77
233 rdf:type schema:Person
234 sg:person.01126227511.07 schema:affiliation grid-institutes:grid.415886.6
235 schema:familyName Grbic
236 schema:givenName Sasa
237 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01126227511.07
238 rdf:type schema:Person
239 sg:person.011402347273.82 schema:affiliation grid-institutes:grid.63368.38
240 schema:familyName Gupta
241 schema:givenName Nakul
242 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011402347273.82
243 rdf:type schema:Person
244 sg:person.011735325401.24 schema:affiliation grid-institutes:grid.415886.6
245 schema:familyName Yoo
246 schema:givenName Youngjin
247 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011735325401.24
248 rdf:type schema:Person
249 sg:person.012162673120.78 schema:affiliation grid-institutes:grid.415886.6
250 schema:familyName Re
251 schema:givenName Thomas
252 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012162673120.78
253 rdf:type schema:Person
254 sg:person.01225416107.94 schema:affiliation grid-institutes:grid.412541.7
255 schema:familyName Nicolaou
256 schema:givenName Savvas
257 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01225416107.94
258 rdf:type schema:Person
259 sg:person.012532706001.83 schema:affiliation grid-institutes:grid.481749.7
260 schema:familyName Ziebandt
261 schema:givenName Valentin
262 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012532706001.83
263 rdf:type schema:Person
264 sg:person.01320676335.36 schema:affiliation grid-institutes:grid.412701.1
265 schema:familyName Barbosa
266 schema:givenName Eduardo J. Mortani
267 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01320676335.36
268 rdf:type schema:Person
269 sg:person.01366452774.22 schema:affiliation grid-institutes:grid.414106.6
270 schema:familyName Mellot
271 schema:givenName François
272 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01366452774.22
273 rdf:type schema:Person
274 sg:person.013722107237.15 schema:affiliation grid-institutes:None
275 schema:familyName Chabin
276 schema:givenName Guillaume
277 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013722107237.15
278 rdf:type schema:Person
279 sg:person.016306171101.99 schema:affiliation grid-institutes:grid.411730.0
280 schema:familyName Aleman
281 schema:givenName Gorka Bastarrika
282 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016306171101.99
283 rdf:type schema:Person
284 sg:person.016574365001.55 schema:affiliation grid-institutes:grid.415886.6
285 schema:familyName Chaganti
286 schema:givenName Shikha
287 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016574365001.55
288 rdf:type schema:Person
289 sg:person.0600344727.20 schema:affiliation grid-institutes:grid.416477.7
290 schema:familyName Sanelli
291 schema:givenName Pina
292 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0600344727.20
293 rdf:type schema:Person
294 sg:person.0655016751.47 schema:affiliation grid-institutes:grid.410567.1
295 schema:familyName Sauter
296 schema:givenName Alexander W.
297 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0655016751.47
298 rdf:type schema:Person
299 sg:person.0703547214.37 schema:affiliation grid-institutes:grid.415886.6
300 schema:familyName Georgescu
301 schema:givenName Bogdan
302 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0703547214.37
303 rdf:type schema:Person
304 sg:person.0754131024.61 schema:affiliation grid-institutes:grid.481749.7
305 schema:familyName Flohr
306 schema:givenName Thomas
307 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0754131024.61
308 rdf:type schema:Person
309 sg:pub.10.1007/s00330-020-07273-y schema:sameAs https://app.dimensions.ai/details/publication/pub.1131095776
310 https://doi.org/10.1007/s00330-020-07273-y
311 rdf:type schema:CreativeWork
312 sg:pub.10.1007/s10096-020-03901-z schema:sameAs https://app.dimensions.ai/details/publication/pub.1127150720
313 https://doi.org/10.1007/s10096-020-03901-z
314 rdf:type schema:CreativeWork
315 sg:pub.10.1038/s41467-020-17971-2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1130092582
316 https://doi.org/10.1038/s41467-020-17971-2
317 rdf:type schema:CreativeWork
318 grid-institutes:None schema:alternateName Health Time, Jaén, Spain
319 Siemens Healthineers, Paris, France
320 schema:name Health Time, Jaén, Spain
321 Siemens Healthineers, Paris, France
322 rdf:type schema:Organization
323 grid-institutes:grid.410567.1 schema:alternateName University Hospital Basel, Clinic of Radiology & Nuclear medicine, Basel, Switzerland
324 schema:name University Hospital Basel, Clinic of Radiology & Nuclear medicine, Basel, Switzerland
325 rdf:type schema:Organization
326 grid-institutes:grid.411730.0 schema:alternateName Clínica Universidad de Navarra, Navarra, Spain
327 schema:name Clínica Universidad de Navarra, Navarra, Spain
328 rdf:type schema:Organization
329 grid-institutes:grid.412541.7 schema:alternateName Vancouver General Hospital, Vancouver, BC Canada
330 schema:name Vancouver General Hospital, Vancouver, BC Canada
331 rdf:type schema:Organization
332 grid-institutes:grid.412701.1 schema:alternateName Penn Medicine, Philadelphia, PA USA
333 schema:name Division of Cardiothoracic Imaging, Department of Radiology, Perelman School of Medicine, University of Pennsylvania, 3400 Spruce Street, Ground Floor Founders Bldg, Philadelphia, PA 19104 USA
334 Penn Medicine, Philadelphia, PA USA
335 rdf:type schema:Organization
336 grid-institutes:grid.414106.6 schema:alternateName Hôpital Foch, Suresnes, France
337 schema:name Hôpital Foch, Suresnes, France
338 rdf:type schema:Organization
339 grid-institutes:grid.415886.6 schema:alternateName Siemens Healthineers, Princeton, NJ USA
340 schema:name Siemens Healthineers, Princeton, NJ USA
341 rdf:type schema:Organization
342 grid-institutes:grid.416477.7 schema:alternateName Donald and Barbara Zucker School of Medicine, Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY USA
343 schema:name Donald and Barbara Zucker School of Medicine, Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY USA
344 rdf:type schema:Organization
345 grid-institutes:grid.481749.7 schema:alternateName Siemens Healthineers, Forchheim, Germany
346 schema:name Siemens Healthineers, Forchheim, Germany
347 rdf:type schema:Organization
348 grid-institutes:grid.63368.38 schema:alternateName Houston Methodist, Houston, TX USA
349 schema:name Houston Methodist, Houston, TX USA
350 rdf:type schema:Organization
 




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


...