Automated segmentation of the fractured vertebrae on CT and its applicability in a radiomics model to predict fracture malignancy View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2022-04-25

AUTHORS

Taeyong Park, Min A Yoon, Young Chul Cho, Su Jung Ham, Yousun Ko, Sehee Kim, Heeryeol Jeong, Jeongjin Lee

ABSTRACT

Although CT radiomics has shown promising results in the evaluation of vertebral fractures, the need for manual segmentation of fractured vertebrae limited the routine clinical implementation of radiomics. Therefore, automated segmentation of fractured vertebrae is needed for successful clinical use of radiomics. In this study, we aimed to develop and validate an automated algorithm for segmentation of fractured vertebral bodies on CT, and to evaluate the applicability of the algorithm in a radiomics prediction model to differentiate benign and malignant fractures. A convolutional neural network was trained to perform automated segmentation of fractured vertebral bodies using 341 vertebrae with benign or malignant fractures from 158 patients, and was validated on independent test sets (internal test, 86 vertebrae [59 patients]; external test, 102 vertebrae [59 patients]). Then, a radiomics model predicting fracture malignancy on CT was constructed, and the prediction performance was compared between automated and human expert segmentations. The algorithm achieved good agreement with human expert segmentation at testing (Dice similarity coefficient, 0.93–0.94; cross-sectional area error, 2.66–2.97%; average surface distance, 0.40–0.54 mm). The radiomics model demonstrated good performance in the training set (AUC, 0.93). In the test sets, automated and human expert segmentations showed comparable prediction performances (AUC, internal test, 0.80 vs 0.87, p = 0.044; external test, 0.83 vs 0.80, p = 0.37). In summary, we developed and validated an automated segmentation algorithm that showed comparable performance to human expert segmentation in a CT radiomics model to predict fracture malignancy, which may enable more practical clinical utilization of radiomics. More... »

PAGES

6735

References to SciGraph publications

  • 2016-07-01. Multi-atlas Segmentation with Joint Label Fusion of Osteoporotic Vertebral Compression Fractures on CT in COMPUTATIONAL METHODS AND CLINICAL APPLICATIONS FOR SPINE IMAGING
  • 2020-05-01. How to develop a meaningful radiomic signature for clinical use in oncologic patients in CANCER IMAGING
  • 2020-07-29. Radiomics feature reproducibility under inter-rater variability in segmentations of CT images in SCIENTIFIC REPORTS
  • 2018-12-05. Vertebral body insufficiency fractures: detection of vertebrae at risk on standard CT images using texture analysis and machine learning in EUROPEAN RADIOLOGY
  • 2019-07-26. Clinical impact of variability on CT radiomics and suggestions for suitable feature selection: a focus on lung cancer in CANCER IMAGING
  • 2015-11-18. U-Net: Convolutional Networks for Biomedical Image Segmentation in MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION – MICCAI 2015
  • 2020-08-12. Radiomics in medical imaging—“how-to” guide and critical reflection in INSIGHTS INTO IMAGING
  • 2019-04-22. A Region-Based Deep Level Set Formulation for Vertebral Bone Segmentation of Osteoporotic Fractures in JOURNAL OF DIGITAL IMAGING
  • 2018-10-30. Automatic segmentation of vertebrae in 3D CT images using adaptive fast 3D pulse coupled neural networks in PHYSICAL AND ENGINEERING SCIENCES IN MEDICINE
  • 2021-03-19. Combined radiomics-clinical model to predict malignancy of vertebral compression fractures on CT in EUROPEAN RADIOLOGY
  • 2014-07-12. A Robust Segmentation Framework for Spine Trauma Diagnosis in COMPUTATIONAL METHODS AND CLINICAL APPLICATIONS FOR SPINE IMAGING
  • 2019-12-21. The effects of volume of interest delineation on MRI-based radiomics analysis: evaluation with two disease groups in CANCER IMAGING
  • 2020-02-01. Metastatic Vertebrae Segmentation for Use in a Clinical Pipeline in COMPUTATIONAL METHODS AND CLINICAL APPLICATIONS FOR SPINE IMAGING
  • 2018-11-14. Radiomics: the facts and the challenges of image analysis in EUROPEAN RADIOLOGY EXPERIMENTAL
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1038/s41598-022-10807-7

    DOI

    http://dx.doi.org/10.1038/s41598-022-10807-7

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1147362821

    PUBMED

    https://www.ncbi.nlm.nih.gov/pubmed/35468985


    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": "Humans", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Neoplasms", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Neural Networks, Computer", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Spinal Fractures", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Spine", 
            "type": "DefinedTerm"
          }, 
          {
            "inDefinedTermSet": "https://www.nlm.nih.gov/mesh/", 
            "name": "Tomography, X-Ray Computed", 
            "type": "DefinedTerm"
          }
        ], 
        "author": [
          {
            "affiliation": {
              "alternateName": "Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.413967.e", 
              "name": [
                "Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Park", 
            "givenName": "Taeyong", 
            "id": "sg:person.016160635401.76", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016160635401.76"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.413967.e", 
              "name": [
                "Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Yoon", 
            "givenName": "Min A", 
            "id": "sg:person.01317564523.14", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01317564523.14"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Biomedical Research Center, Asan Institute for Life Sciences, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.413967.e", 
              "name": [
                "Biomedical Research Center, Asan Institute for Life Sciences, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Cho", 
            "givenName": "Young Chul", 
            "id": "sg:person.01324133120.03", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01324133120.03"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.413967.e", 
              "name": [
                "Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Ham", 
            "givenName": "Su Jung", 
            "id": "sg:person.0576122127.23", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0576122127.23"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Biomedical Research Center, Asan Institute for Life Sciences, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.413967.e", 
              "name": [
                "Biomedical Research Center, Asan Institute for Life Sciences, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Ko", 
            "givenName": "Yousun", 
            "id": "sg:person.01137623345.18", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01137623345.18"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.413967.e", 
              "name": [
                "Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Kim", 
            "givenName": "Sehee", 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "School of Computer Science and Engineering, Soongsil University, 369 Sangdo-ro, Dongjak-gu, 156-743, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.263765.3", 
              "name": [
                "School of Computer Science and Engineering, Soongsil University, 369 Sangdo-ro, Dongjak-gu, 156-743, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Jeong", 
            "givenName": "Heeryeol", 
            "id": "sg:person.014001312403.46", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014001312403.46"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "School of Computer Science and Engineering, Soongsil University, 369 Sangdo-ro, Dongjak-gu, 156-743, Seoul, Korea", 
              "id": "http://www.grid.ac/institutes/grid.263765.3", 
              "name": [
                "School of Computer Science and Engineering, Soongsil University, 369 Sangdo-ro, Dongjak-gu, 156-743, Seoul, Korea"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Lee", 
            "givenName": "Jeongjin", 
            "id": "sg:person.01011505663.09", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01011505663.09"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1186/s41747-018-0068-z", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1109834161", 
              "https://doi.org/10.1186/s41747-018-0068-z"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1186/s40644-020-00311-4", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1127313755", 
              "https://doi.org/10.1186/s40644-020-00311-4"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s00330-018-5846-8", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1110398307", 
              "https://doi.org/10.1007/s00330-018-5846-8"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s13246-018-0702-3", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1107924748", 
              "https://doi.org/10.1007/s13246-018-0702-3"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s10278-019-00216-0", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1113623335", 
              "https://doi.org/10.1007/s10278-019-00216-0"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1186/s40644-019-0276-7", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1123595070", 
              "https://doi.org/10.1186/s40644-019-0276-7"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1186/s13244-020-00887-2", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1130076141", 
              "https://doi.org/10.1186/s13244-020-00887-2"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1186/s40644-019-0239-z", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1119864713", 
              "https://doi.org/10.1186/s40644-019-0239-z"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-319-07269-2_3", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1006670394", 
              "https://doi.org/10.1007/978-3-319-07269-2_3"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-319-24574-4_28", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1017774818", 
              "https://doi.org/10.1007/978-3-319-24574-4_28"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-030-39752-4_2", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1124458915", 
              "https://doi.org/10.1007/978-3-030-39752-4_2"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1038/s41598-020-69534-6", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1129710367", 
              "https://doi.org/10.1038/s41598-020-69534-6"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s00330-021-07832-x", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1136537140", 
              "https://doi.org/10.1007/s00330-021-07832-x"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-319-41827-8_7", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1019423124", 
              "https://doi.org/10.1007/978-3-319-41827-8_7"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2022-04-25", 
        "datePublishedReg": "2022-04-25", 
        "description": "Although CT radiomics has shown promising results in the evaluation of vertebral fractures, the need for manual segmentation of fractured vertebrae limited the routine clinical implementation of radiomics. Therefore, automated segmentation of fractured vertebrae is needed for successful clinical use of radiomics. In this study, we aimed to develop and validate an automated algorithm for segmentation of fractured vertebral bodies on CT, and to evaluate the applicability of the algorithm in a radiomics prediction model to differentiate benign and malignant fractures. A convolutional neural network was trained to perform automated segmentation of fractured vertebral bodies using 341 vertebrae with benign or malignant fractures from 158 patients, and was validated on independent test sets (internal test, 86 vertebrae [59 patients]; external test, 102 vertebrae [59 patients]). Then, a radiomics model predicting fracture malignancy on CT was constructed, and the prediction performance was compared between automated and human expert segmentations. The algorithm achieved good agreement with human expert segmentation at testing (Dice similarity coefficient, 0.93\u20130.94; cross-sectional area error, 2.66\u20132.97%; average surface distance, 0.40\u20130.54\u00a0mm). The radiomics model demonstrated good performance in the training set (AUC, 0.93). In the test sets, automated and human expert segmentations showed comparable prediction performances (AUC, internal test, 0.80 vs 0.87, p\u2009=\u20090.044; external test, 0.83 vs 0.80, p\u2009=\u20090.37). In summary, we developed and validated an automated segmentation algorithm that showed comparable performance to human expert segmentation in a CT radiomics model to predict fracture malignancy, which may enable more practical clinical utilization of radiomics.", 
        "genre": "article", 
        "id": "sg:pub.10.1038/s41598-022-10807-7", 
        "isAccessibleForFree": true, 
        "isPartOf": [
          {
            "id": "sg:journal.1045337", 
            "issn": [
              "2045-2322"
            ], 
            "name": "Scientific Reports", 
            "publisher": "Springer Nature", 
            "type": "Periodical"
          }, 
          {
            "issueNumber": "1", 
            "type": "PublicationIssue"
          }, 
          {
            "type": "PublicationVolume", 
            "volumeNumber": "12"
          }
        ], 
        "keywords": [
          "fractured vertebral body", 
          "fractured vertebra", 
          "radiomics model", 
          "malignant fractures", 
          "vertebral body", 
          "radiomics prediction model", 
          "successful clinical use", 
          "vertebral fractures", 
          "routine clinical implementation", 
          "CT radiomics", 
          "clinical use", 
          "clinical utilization", 
          "malignancy", 
          "human expert segmentation", 
          "clinical implementation", 
          "CT", 
          "radiomics", 
          "vertebrae", 
          "expert segmentation", 
          "fractures", 
          "patients", 
          "convolutional neural network", 
          "prediction performance", 
          "promising results", 
          "test set", 
          "comparable prediction performance", 
          "neural network", 
          "segmentation algorithm", 
          "manual segmentation", 
          "segmentation", 
          "body", 
          "independent test set", 
          "training set", 
          "algorithm", 
          "comparable performance", 
          "testing", 
          "better performance", 
          "evaluation", 
          "summary", 
          "study", 
          "prediction model", 
          "need", 
          "use", 
          "set", 
          "performance", 
          "model", 
          "utilization", 
          "network", 
          "results", 
          "implementation", 
          "applicability", 
          "agreement", 
          "good agreement"
        ], 
        "name": "Automated segmentation of the fractured vertebrae on CT and its applicability in a radiomics model to predict fracture malignancy", 
        "pagination": "6735", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1147362821"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1038/s41598-022-10807-7"
            ]
          }, 
          {
            "name": "pubmed_id", 
            "type": "PropertyValue", 
            "value": [
              "35468985"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1038/s41598-022-10807-7", 
          "https://app.dimensions.ai/details/publication/pub.1147362821"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2022-09-02T16:07", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-springernature-scigraph/baseset/20220902/entities/gbq_results/article/article_941.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://doi.org/10.1038/s41598-022-10807-7"
      }
    ]
     

    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/s41598-022-10807-7'

    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/s41598-022-10807-7'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1038/s41598-022-10807-7'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1038/s41598-022-10807-7'


     

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

    248 TRIPLES      21 PREDICATES      98 URIs      76 LITERALS      13 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1038/s41598-022-10807-7 schema:about N3912a71845bf477dab2df7f749887283
    2 N76ef59e58a2143338bf5ae709a1cf067
    3 N786f1002ffc346b1824b62d830cc934c
    4 Ndd62c0e7a0fb43efba1c52a035fcbb47
    5 Ne1fd9a2a8d75446ca686bd98299eb193
    6 Nfa707bf0ba604cf8926f303249989a3f
    7 anzsrc-for:11
    8 anzsrc-for:1103
    9 schema:author Nbad2ce5e7a0f49d795eb4251cf4fbaf5
    10 schema:citation sg:pub.10.1007/978-3-030-39752-4_2
    11 sg:pub.10.1007/978-3-319-07269-2_3
    12 sg:pub.10.1007/978-3-319-24574-4_28
    13 sg:pub.10.1007/978-3-319-41827-8_7
    14 sg:pub.10.1007/s00330-018-5846-8
    15 sg:pub.10.1007/s00330-021-07832-x
    16 sg:pub.10.1007/s10278-019-00216-0
    17 sg:pub.10.1007/s13246-018-0702-3
    18 sg:pub.10.1038/s41598-020-69534-6
    19 sg:pub.10.1186/s13244-020-00887-2
    20 sg:pub.10.1186/s40644-019-0239-z
    21 sg:pub.10.1186/s40644-019-0276-7
    22 sg:pub.10.1186/s40644-020-00311-4
    23 sg:pub.10.1186/s41747-018-0068-z
    24 schema:datePublished 2022-04-25
    25 schema:datePublishedReg 2022-04-25
    26 schema:description Although CT radiomics has shown promising results in the evaluation of vertebral fractures, the need for manual segmentation of fractured vertebrae limited the routine clinical implementation of radiomics. Therefore, automated segmentation of fractured vertebrae is needed for successful clinical use of radiomics. In this study, we aimed to develop and validate an automated algorithm for segmentation of fractured vertebral bodies on CT, and to evaluate the applicability of the algorithm in a radiomics prediction model to differentiate benign and malignant fractures. A convolutional neural network was trained to perform automated segmentation of fractured vertebral bodies using 341 vertebrae with benign or malignant fractures from 158 patients, and was validated on independent test sets (internal test, 86 vertebrae [59 patients]; external test, 102 vertebrae [59 patients]). Then, a radiomics model predicting fracture malignancy on CT was constructed, and the prediction performance was compared between automated and human expert segmentations. The algorithm achieved good agreement with human expert segmentation at testing (Dice similarity coefficient, 0.93–0.94; cross-sectional area error, 2.66–2.97%; average surface distance, 0.40–0.54 mm). The radiomics model demonstrated good performance in the training set (AUC, 0.93). In the test sets, automated and human expert segmentations showed comparable prediction performances (AUC, internal test, 0.80 vs 0.87, p = 0.044; external test, 0.83 vs 0.80, p = 0.37). In summary, we developed and validated an automated segmentation algorithm that showed comparable performance to human expert segmentation in a CT radiomics model to predict fracture malignancy, which may enable more practical clinical utilization of radiomics.
    27 schema:genre article
    28 schema:isAccessibleForFree true
    29 schema:isPartOf Nd2dfb872204642288e67739fed3ad38f
    30 Nff4f7a09e10046e49884747605824cf2
    31 sg:journal.1045337
    32 schema:keywords CT
    33 CT radiomics
    34 agreement
    35 algorithm
    36 applicability
    37 better performance
    38 body
    39 clinical implementation
    40 clinical use
    41 clinical utilization
    42 comparable performance
    43 comparable prediction performance
    44 convolutional neural network
    45 evaluation
    46 expert segmentation
    47 fractured vertebra
    48 fractured vertebral body
    49 fractures
    50 good agreement
    51 human expert segmentation
    52 implementation
    53 independent test set
    54 malignancy
    55 malignant fractures
    56 manual segmentation
    57 model
    58 need
    59 network
    60 neural network
    61 patients
    62 performance
    63 prediction model
    64 prediction performance
    65 promising results
    66 radiomics
    67 radiomics model
    68 radiomics prediction model
    69 results
    70 routine clinical implementation
    71 segmentation
    72 segmentation algorithm
    73 set
    74 study
    75 successful clinical use
    76 summary
    77 test set
    78 testing
    79 training set
    80 use
    81 utilization
    82 vertebrae
    83 vertebral body
    84 vertebral fractures
    85 schema:name Automated segmentation of the fractured vertebrae on CT and its applicability in a radiomics model to predict fracture malignancy
    86 schema:pagination 6735
    87 schema:productId N291828e2be004406936c55c9a472116f
    88 N366c9f23bf4f49de9dde5be4127bde88
    89 Ne94a603000424298a849188ac689ffa3
    90 schema:sameAs https://app.dimensions.ai/details/publication/pub.1147362821
    91 https://doi.org/10.1038/s41598-022-10807-7
    92 schema:sdDatePublished 2022-09-02T16:07
    93 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    94 schema:sdPublisher Nfe81f4df762444468ca7775b9aef9c1f
    95 schema:url https://doi.org/10.1038/s41598-022-10807-7
    96 sgo:license sg:explorer/license/
    97 sgo:sdDataset articles
    98 rdf:type schema:ScholarlyArticle
    99 N200570fcc56741778225aef552eb1d99 rdf:first Nfa65ceb95ace4550b2229c96541650f4
    100 rdf:rest N946fc3b97e8d4646bee95d58a8b36964
    101 N291828e2be004406936c55c9a472116f schema:name dimensions_id
    102 schema:value pub.1147362821
    103 rdf:type schema:PropertyValue
    104 N366c9f23bf4f49de9dde5be4127bde88 schema:name pubmed_id
    105 schema:value 35468985
    106 rdf:type schema:PropertyValue
    107 N3912a71845bf477dab2df7f749887283 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    108 schema:name Neoplasms
    109 rdf:type schema:DefinedTerm
    110 N3a35ff88a06c4abb939a9f136b942104 rdf:first sg:person.0576122127.23
    111 rdf:rest N4549d3d805a4459b95fb5ad66f12f10d
    112 N4549d3d805a4459b95fb5ad66f12f10d rdf:first sg:person.01137623345.18
    113 rdf:rest N200570fcc56741778225aef552eb1d99
    114 N76ef59e58a2143338bf5ae709a1cf067 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    115 schema:name Spine
    116 rdf:type schema:DefinedTerm
    117 N786f1002ffc346b1824b62d830cc934c schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    118 schema:name Tomography, X-Ray Computed
    119 rdf:type schema:DefinedTerm
    120 N946fc3b97e8d4646bee95d58a8b36964 rdf:first sg:person.014001312403.46
    121 rdf:rest Ne7585ab8c8854d5283f0e8873e88db0f
    122 Nb88dbbb8bb0b49b098fe96231fbb8815 rdf:first sg:person.01317564523.14
    123 rdf:rest Nfed3be2c744d423889291184740307e5
    124 Nbad2ce5e7a0f49d795eb4251cf4fbaf5 rdf:first sg:person.016160635401.76
    125 rdf:rest Nb88dbbb8bb0b49b098fe96231fbb8815
    126 Nd2dfb872204642288e67739fed3ad38f schema:volumeNumber 12
    127 rdf:type schema:PublicationVolume
    128 Ndd62c0e7a0fb43efba1c52a035fcbb47 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    129 schema:name Humans
    130 rdf:type schema:DefinedTerm
    131 Ne1fd9a2a8d75446ca686bd98299eb193 schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    132 schema:name Spinal Fractures
    133 rdf:type schema:DefinedTerm
    134 Ne7585ab8c8854d5283f0e8873e88db0f rdf:first sg:person.01011505663.09
    135 rdf:rest rdf:nil
    136 Ne94a603000424298a849188ac689ffa3 schema:name doi
    137 schema:value 10.1038/s41598-022-10807-7
    138 rdf:type schema:PropertyValue
    139 Nfa65ceb95ace4550b2229c96541650f4 schema:affiliation grid-institutes:grid.413967.e
    140 schema:familyName Kim
    141 schema:givenName Sehee
    142 rdf:type schema:Person
    143 Nfa707bf0ba604cf8926f303249989a3f schema:inDefinedTermSet https://www.nlm.nih.gov/mesh/
    144 schema:name Neural Networks, Computer
    145 rdf:type schema:DefinedTerm
    146 Nfe81f4df762444468ca7775b9aef9c1f schema:name Springer Nature - SN SciGraph project
    147 rdf:type schema:Organization
    148 Nfed3be2c744d423889291184740307e5 rdf:first sg:person.01324133120.03
    149 rdf:rest N3a35ff88a06c4abb939a9f136b942104
    150 Nff4f7a09e10046e49884747605824cf2 schema:issueNumber 1
    151 rdf:type schema:PublicationIssue
    152 anzsrc-for:11 schema:inDefinedTermSet anzsrc-for:
    153 schema:name Medical and Health Sciences
    154 rdf:type schema:DefinedTerm
    155 anzsrc-for:1103 schema:inDefinedTermSet anzsrc-for:
    156 schema:name Clinical Sciences
    157 rdf:type schema:DefinedTerm
    158 sg:journal.1045337 schema:issn 2045-2322
    159 schema:name Scientific Reports
    160 schema:publisher Springer Nature
    161 rdf:type schema:Periodical
    162 sg:person.01011505663.09 schema:affiliation grid-institutes:grid.263765.3
    163 schema:familyName Lee
    164 schema:givenName Jeongjin
    165 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01011505663.09
    166 rdf:type schema:Person
    167 sg:person.01137623345.18 schema:affiliation grid-institutes:grid.413967.e
    168 schema:familyName Ko
    169 schema:givenName Yousun
    170 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01137623345.18
    171 rdf:type schema:Person
    172 sg:person.01317564523.14 schema:affiliation grid-institutes:grid.413967.e
    173 schema:familyName Yoon
    174 schema:givenName Min A
    175 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01317564523.14
    176 rdf:type schema:Person
    177 sg:person.01324133120.03 schema:affiliation grid-institutes:grid.413967.e
    178 schema:familyName Cho
    179 schema:givenName Young Chul
    180 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01324133120.03
    181 rdf:type schema:Person
    182 sg:person.014001312403.46 schema:affiliation grid-institutes:grid.263765.3
    183 schema:familyName Jeong
    184 schema:givenName Heeryeol
    185 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.014001312403.46
    186 rdf:type schema:Person
    187 sg:person.016160635401.76 schema:affiliation grid-institutes:grid.413967.e
    188 schema:familyName Park
    189 schema:givenName Taeyong
    190 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016160635401.76
    191 rdf:type schema:Person
    192 sg:person.0576122127.23 schema:affiliation grid-institutes:grid.413967.e
    193 schema:familyName Ham
    194 schema:givenName Su Jung
    195 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0576122127.23
    196 rdf:type schema:Person
    197 sg:pub.10.1007/978-3-030-39752-4_2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1124458915
    198 https://doi.org/10.1007/978-3-030-39752-4_2
    199 rdf:type schema:CreativeWork
    200 sg:pub.10.1007/978-3-319-07269-2_3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006670394
    201 https://doi.org/10.1007/978-3-319-07269-2_3
    202 rdf:type schema:CreativeWork
    203 sg:pub.10.1007/978-3-319-24574-4_28 schema:sameAs https://app.dimensions.ai/details/publication/pub.1017774818
    204 https://doi.org/10.1007/978-3-319-24574-4_28
    205 rdf:type schema:CreativeWork
    206 sg:pub.10.1007/978-3-319-41827-8_7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1019423124
    207 https://doi.org/10.1007/978-3-319-41827-8_7
    208 rdf:type schema:CreativeWork
    209 sg:pub.10.1007/s00330-018-5846-8 schema:sameAs https://app.dimensions.ai/details/publication/pub.1110398307
    210 https://doi.org/10.1007/s00330-018-5846-8
    211 rdf:type schema:CreativeWork
    212 sg:pub.10.1007/s00330-021-07832-x schema:sameAs https://app.dimensions.ai/details/publication/pub.1136537140
    213 https://doi.org/10.1007/s00330-021-07832-x
    214 rdf:type schema:CreativeWork
    215 sg:pub.10.1007/s10278-019-00216-0 schema:sameAs https://app.dimensions.ai/details/publication/pub.1113623335
    216 https://doi.org/10.1007/s10278-019-00216-0
    217 rdf:type schema:CreativeWork
    218 sg:pub.10.1007/s13246-018-0702-3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1107924748
    219 https://doi.org/10.1007/s13246-018-0702-3
    220 rdf:type schema:CreativeWork
    221 sg:pub.10.1038/s41598-020-69534-6 schema:sameAs https://app.dimensions.ai/details/publication/pub.1129710367
    222 https://doi.org/10.1038/s41598-020-69534-6
    223 rdf:type schema:CreativeWork
    224 sg:pub.10.1186/s13244-020-00887-2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1130076141
    225 https://doi.org/10.1186/s13244-020-00887-2
    226 rdf:type schema:CreativeWork
    227 sg:pub.10.1186/s40644-019-0239-z schema:sameAs https://app.dimensions.ai/details/publication/pub.1119864713
    228 https://doi.org/10.1186/s40644-019-0239-z
    229 rdf:type schema:CreativeWork
    230 sg:pub.10.1186/s40644-019-0276-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1123595070
    231 https://doi.org/10.1186/s40644-019-0276-7
    232 rdf:type schema:CreativeWork
    233 sg:pub.10.1186/s40644-020-00311-4 schema:sameAs https://app.dimensions.ai/details/publication/pub.1127313755
    234 https://doi.org/10.1186/s40644-020-00311-4
    235 rdf:type schema:CreativeWork
    236 sg:pub.10.1186/s41747-018-0068-z schema:sameAs https://app.dimensions.ai/details/publication/pub.1109834161
    237 https://doi.org/10.1186/s41747-018-0068-z
    238 rdf:type schema:CreativeWork
    239 grid-institutes:grid.263765.3 schema:alternateName School of Computer Science and Engineering, Soongsil University, 369 Sangdo-ro, Dongjak-gu, 156-743, Seoul, Korea
    240 schema:name School of Computer Science and Engineering, Soongsil University, 369 Sangdo-ro, Dongjak-gu, 156-743, Seoul, Korea
    241 rdf:type schema:Organization
    242 grid-institutes:grid.413967.e schema:alternateName Biomedical Research Center, Asan Institute for Life Sciences, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea
    243 Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea
    244 Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea
    245 schema:name Biomedical Research Center, Asan Institute for Life Sciences, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea
    246 Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea
    247 Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-ro 43-gil, Songpa-gu, 05505, Seoul, Korea
    248 rdf:type schema:Organization
     




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


    ...