Fully convolutional architecture vs sliding-window CNN for corneal endothelium cell segmentation View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2019-12

AUTHORS

Juan P. Vigueras-Guillén, Busra Sari, Stanley F. Goes, Hans G. Lemij, Jeroen van Rooij, Koenraad A. Vermeer, Lucas J. van Vliet

ABSTRACT

Corneal endothelium (CE) images provide valuable clinical information regarding the health state of the cornea. Computation of the clinical morphometric parameters requires the segmentation of endothelial cell images. Current techniques to image the endothelium in vivo deliver low quality images, which makes automatic segmentation a complicated task. Here, we present two convolutional neural networks (CNN) to segment CE images: a global fully convolutional approach based on U-net, and a local sliding-window network (SW-net). We propose to use probabilistic labels instead of binary, we evaluate a preprocessing method to enhance the contrast of images, and we introduce a postprocessing method based on Fourier analysis and watershed to convert the CNN output images into the final cell segmentation. Both methods are applied to 50 images acquired with an SP-1P Topcon specular microscope. Estimates are compared against a manual delineation made by a trained observer. U-net (AUC=0.9938) yields slightly sharper, clearer images than SW-net (AUC=0.9921). After postprocessing, U-net obtains a DICE=0.981 and a MHD=0.22 (modified Hausdorff distance), whereas SW-net yields a DICE=0.978 and a MHD=0.30. U-net generates a wrong cell segmentation in only 0.48% of the cells, versus 0.92% for the SW-net. U-net achieves statistically significant better precision and accuracy than both, Topcon and SW-net, for the estimates of three clinical parameters: cell density (ECD), polymegethism (CV), and pleomorphism (HEX). The mean relative error in U-net for the parameters is 0.4% in ECD, 2.8% in CV, and 1.3% in HEX. The computation time to segment an image and estimate the parameters is barely a few seconds. Both methods presented here provide a statistically significant improvement over the state of the art. U-net has reached the smallest error rate. We suggest a segmentation refinement based on our previous work to further improve the performance. More... »

PAGES

4

References to SciGraph publications

Identifiers

URI

http://scigraph.springernature.com/pub.10.1186/s42490-019-0003-2

DOI

http://dx.doi.org/10.1186/s42490-019-0003-2

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1111775678


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/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "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"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Rotterdam Eye Hospital", 
          "id": "https://www.grid.ac/institutes/grid.414699.7", 
          "name": [
            "Delft University of Technology, Dept. of Imaging Physics, Lorentzweg 1, 2628CJ, Delft, The Netherlands", 
            "Rotterdam Ophthalmic Institute, Schiedamse Vest 160, 3011BH, Rotterdam, The Netherlands"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Vigueras-Guill\u00e9n", 
        "givenName": "Juan P.", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Delft University of Technology", 
          "id": "https://www.grid.ac/institutes/grid.5292.c", 
          "name": [
            "Delft University of Technology, Dept. of Imaging Physics, Lorentzweg 1, 2628CJ, Delft, The Netherlands"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Sari", 
        "givenName": "Busra", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Delft University of Technology", 
          "id": "https://www.grid.ac/institutes/grid.5292.c", 
          "name": [
            "Delft University of Technology, Dept. of Imaging Physics, Lorentzweg 1, 2628CJ, Delft, The Netherlands"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Goes", 
        "givenName": "Stanley F.", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Rotterdam Eye Hospital", 
          "id": "https://www.grid.ac/institutes/grid.414699.7", 
          "name": [
            "The Rotterdam Eye Hospital, Schiedamse Vest 180, 3011BH, Rotterdam, The Netherlands"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Lemij", 
        "givenName": "Hans G.", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Rotterdam Eye Hospital", 
          "id": "https://www.grid.ac/institutes/grid.414699.7", 
          "name": [
            "The Rotterdam Eye Hospital, Schiedamse Vest 180, 3011BH, Rotterdam, The Netherlands"
          ], 
          "type": "Organization"
        }, 
        "familyName": "van Rooij", 
        "givenName": "Jeroen", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Rotterdam Eye Hospital", 
          "id": "https://www.grid.ac/institutes/grid.414699.7", 
          "name": [
            "Rotterdam Ophthalmic Institute, Schiedamse Vest 160, 3011BH, Rotterdam, The Netherlands"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Vermeer", 
        "givenName": "Koenraad A.", 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Delft University of Technology", 
          "id": "https://www.grid.ac/institutes/grid.5292.c", 
          "name": [
            "Delft University of Technology, Dept. of Imaging Physics, Lorentzweg 1, 2628CJ, Delft, The Netherlands"
          ], 
          "type": "Organization"
        }, 
        "familyName": "van Vliet", 
        "givenName": "Lucas J.", 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1186/s12880-015-0054-3", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1001794110", 
          "https://doi.org/10.1186/s12880-015-0054-3"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/s12880-015-0054-3", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1001794110", 
          "https://doi.org/10.1186/s12880-015-0054-3"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1038/sj.eye.6700559", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1004950511", 
          "https://doi.org/10.1038/sj.eye.6700559"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1038/sj.eye.6700559", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1004950511", 
          "https://doi.org/10.1038/sj.eye.6700559"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-3-319-23814-2_14", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1005182543", 
          "https://doi.org/10.1007/978-3-319-23814-2_14"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.compmedimag.2016.07.010", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1006504630"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0b013e31802be629", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1006805600"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0b013e31802be629", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1006805600"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0b013e31802be629", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1006805600"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1002/jmor.1051740104", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1007940510"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1162/neco.1989.1.4.541", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1008345178"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/00003226-200308000-00005", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1009308127"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/00003226-200308000-00005", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1009308127"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0010-4825(99)00010-4", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1015179935"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/s0031-3203(00)00074-1", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1017163976"
        ], 
        "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": "https://doi.org/10.1016/s0734-189x(87)80186-x", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1019291816"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.ajo.2009.04.016", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1023619767"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.cmpb.2015.09.003", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1025329845"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1027055606", 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4419-0300-6", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1027055606", 
          "https://doi.org/10.1007/978-1-4419-0300-6"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-1-4419-0300-6", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1027055606", 
          "https://doi.org/10.1007/978-1-4419-0300-6"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.3349/ymj.2011.52.2.322", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1033041955"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0000000000000908", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1038391544"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0000000000000908", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1038391544"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0000000000000385", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1039021041"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0000000000000385", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1039021041"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/01.ico.0000151505.03824.6c", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1039134662"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/01.ico.0000151505.03824.6c", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1039134662"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/01.ico.0000151505.03824.6c", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1039134662"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.ajo.2011.10.024", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1043433263"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1136/bjo.2009.166561", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1043676105"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0b013e31825de8fa", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1045412299"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/ico.0b013e31825de8fa", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1045412299"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1117/12.60644", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1050290921"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tfuzz.2016.2574915", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061607167"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tpami.2016.2572683", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1061745111"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.2307/1932409", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1069656769"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.3928/15428877-20110812-04", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1071728620"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.5566/ias.v27.p53-61", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1072988760"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/iembs.2007.4353724", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1077517956"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1078511631", 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://app.dimensions.ai/details/publication/pub.1083061562", 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/icl.0000000000000362", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1084194696"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/icl.0000000000000362", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1084194696"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1097/icl.0000000000000362", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1084194696"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1117/12.2254342", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1084714918"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1117/12.2264430", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1085407403"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/3065386", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1085642448"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1145/3065386", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1085642448"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/cbms.1992.245000", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1086365009"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.23915/distill.00003", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1087157001"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/icpr.1994.576361", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094517408"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/icdar.2003.1227801", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1094714779"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.15439/2017f54", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1095851541"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.3390/sym10030060", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1101350073"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.cmpb.2018.03.015", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1101697241"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.artmed.2018.04.004", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1103422892"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1016/j.artmed.2018.04.004", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1103422892"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tmi.2018.2841910", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1104318661"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.1109/tmi.2018.2841910", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1104318661"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-3-319-93000-8_72", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1104412424", 
          "https://doi.org/10.1007/978-3-319-93000-8_72"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/978-3-319-93000-8_72", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1104412424", 
          "https://doi.org/10.1007/978-3-319-93000-8_72"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2019-12", 
    "datePublishedReg": "2019-12-01", 
    "description": "Corneal endothelium (CE) images provide valuable clinical information regarding the health state of the cornea. Computation of the clinical morphometric parameters requires the segmentation of endothelial cell images. Current techniques to image the endothelium in vivo deliver low quality images, which makes automatic segmentation a complicated task. Here, we present two convolutional neural networks (CNN) to segment CE images: a global fully convolutional approach based on U-net, and a local sliding-window network (SW-net). We propose to use probabilistic labels instead of binary, we evaluate a preprocessing method to enhance the contrast of images, and we introduce a postprocessing method based on Fourier analysis and watershed to convert the CNN output images into the final cell segmentation. Both methods are applied to 50 images acquired with an SP-1P Topcon specular microscope. Estimates are compared against a manual delineation made by a trained observer. U-net (AUC=0.9938) yields slightly sharper, clearer images than SW-net (AUC=0.9921). After postprocessing, U-net obtains a DICE=0.981 and a MHD=0.22 (modified Hausdorff distance), whereas SW-net yields a DICE=0.978 and a MHD=0.30. U-net generates a wrong cell segmentation in only 0.48% of the cells, versus 0.92% for the SW-net. U-net achieves statistically significant better precision and accuracy than both, Topcon and SW-net, for the estimates of three clinical parameters: cell density (ECD), polymegethism (CV), and pleomorphism (HEX). The mean relative error in U-net for the parameters is 0.4% in ECD, 2.8% in CV, and 1.3% in HEX. The computation time to segment an image and estimate the parameters is barely a few seconds. Both methods presented here provide a statistically significant improvement over the state of the art. U-net has reached the smallest error rate. We suggest a segmentation refinement based on our previous work to further improve the performance.", 
    "genre": "research_article", 
    "id": "sg:pub.10.1186/s42490-019-0003-2", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": false, 
    "isPartOf": [
      {
        "id": "sg:journal.1336331", 
        "issn": [
          "2524-4426"
        ], 
        "name": "BMC Biomedical Engineering", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "1", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "1"
      }
    ], 
    "name": "Fully convolutional architecture vs sliding-window CNN for corneal endothelium cell segmentation", 
    "pagination": "4", 
    "productId": [
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "8b1d748b6c91ab573167c17cfc2db048731c72c1bd3b6219142f2221889cc711"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1186/s42490-019-0003-2"
        ]
      }, 
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1111775678"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1186/s42490-019-0003-2", 
      "https://app.dimensions.ai/details/publication/pub.1111775678"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2019-04-11T08:58", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000326_0000000326/records_68444_00000000.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://link.springer.com/10.1186%2Fs42490-019-0003-2"
  }
]
 

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.10.1186/s42490-019-0003-2'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1186/s42490-019-0003-2'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1186/s42490-019-0003-2'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1186/s42490-019-0003-2'


 

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

241 TRIPLES      21 PREDICATES      73 URIs      19 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1186/s42490-019-0003-2 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author N0020a08eb65d4459af7489cc4d3b8e3e
4 schema:citation sg:pub.10.1007/978-1-4419-0300-6
5 sg:pub.10.1007/978-3-319-23814-2_14
6 sg:pub.10.1007/978-3-319-24574-4_28
7 sg:pub.10.1007/978-3-319-93000-8_72
8 sg:pub.10.1038/sj.eye.6700559
9 sg:pub.10.1186/s12880-015-0054-3
10 https://app.dimensions.ai/details/publication/pub.1027055606
11 https://app.dimensions.ai/details/publication/pub.1078511631
12 https://app.dimensions.ai/details/publication/pub.1083061562
13 https://doi.org/10.1002/jmor.1051740104
14 https://doi.org/10.1016/j.ajo.2009.04.016
15 https://doi.org/10.1016/j.ajo.2011.10.024
16 https://doi.org/10.1016/j.artmed.2018.04.004
17 https://doi.org/10.1016/j.cmpb.2015.09.003
18 https://doi.org/10.1016/j.cmpb.2018.03.015
19 https://doi.org/10.1016/j.compmedimag.2016.07.010
20 https://doi.org/10.1016/s0010-4825(99)00010-4
21 https://doi.org/10.1016/s0031-3203(00)00074-1
22 https://doi.org/10.1016/s0734-189x(87)80186-x
23 https://doi.org/10.1097/00003226-200308000-00005
24 https://doi.org/10.1097/01.ico.0000151505.03824.6c
25 https://doi.org/10.1097/icl.0000000000000362
26 https://doi.org/10.1097/ico.0000000000000385
27 https://doi.org/10.1097/ico.0000000000000908
28 https://doi.org/10.1097/ico.0b013e31802be629
29 https://doi.org/10.1097/ico.0b013e31825de8fa
30 https://doi.org/10.1109/cbms.1992.245000
31 https://doi.org/10.1109/icdar.2003.1227801
32 https://doi.org/10.1109/icpr.1994.576361
33 https://doi.org/10.1109/iembs.2007.4353724
34 https://doi.org/10.1109/tfuzz.2016.2574915
35 https://doi.org/10.1109/tmi.2018.2841910
36 https://doi.org/10.1109/tpami.2016.2572683
37 https://doi.org/10.1117/12.2254342
38 https://doi.org/10.1117/12.2264430
39 https://doi.org/10.1117/12.60644
40 https://doi.org/10.1136/bjo.2009.166561
41 https://doi.org/10.1145/3065386
42 https://doi.org/10.1162/neco.1989.1.4.541
43 https://doi.org/10.15439/2017f54
44 https://doi.org/10.2307/1932409
45 https://doi.org/10.23915/distill.00003
46 https://doi.org/10.3349/ymj.2011.52.2.322
47 https://doi.org/10.3390/sym10030060
48 https://doi.org/10.3928/15428877-20110812-04
49 https://doi.org/10.5566/ias.v27.p53-61
50 schema:datePublished 2019-12
51 schema:datePublishedReg 2019-12-01
52 schema:description Corneal endothelium (CE) images provide valuable clinical information regarding the health state of the cornea. Computation of the clinical morphometric parameters requires the segmentation of endothelial cell images. Current techniques to image the endothelium in vivo deliver low quality images, which makes automatic segmentation a complicated task. Here, we present two convolutional neural networks (CNN) to segment CE images: a global fully convolutional approach based on U-net, and a local sliding-window network (SW-net). We propose to use probabilistic labels instead of binary, we evaluate a preprocessing method to enhance the contrast of images, and we introduce a postprocessing method based on Fourier analysis and watershed to convert the CNN output images into the final cell segmentation. Both methods are applied to 50 images acquired with an SP-1P Topcon specular microscope. Estimates are compared against a manual delineation made by a trained observer. U-net (AUC=0.9938) yields slightly sharper, clearer images than SW-net (AUC=0.9921). After postprocessing, U-net obtains a DICE=0.981 and a MHD=0.22 (modified Hausdorff distance), whereas SW-net yields a DICE=0.978 and a MHD=0.30. U-net generates a wrong cell segmentation in only 0.48% of the cells, versus 0.92% for the SW-net. U-net achieves statistically significant better precision and accuracy than both, Topcon and SW-net, for the estimates of three clinical parameters: cell density (ECD), polymegethism (CV), and pleomorphism (HEX). The mean relative error in U-net for the parameters is 0.4% in ECD, 2.8% in CV, and 1.3% in HEX. The computation time to segment an image and estimate the parameters is barely a few seconds. Both methods presented here provide a statistically significant improvement over the state of the art. U-net has reached the smallest error rate. We suggest a segmentation refinement based on our previous work to further improve the performance.
53 schema:genre research_article
54 schema:inLanguage en
55 schema:isAccessibleForFree false
56 schema:isPartOf N39d7eb3182c243f79bf98929876e5cde
57 Na40bb2e9849445cbb07028aa54ba199d
58 sg:journal.1336331
59 schema:name Fully convolutional architecture vs sliding-window CNN for corneal endothelium cell segmentation
60 schema:pagination 4
61 schema:productId N4841d01270cc45ac98aa4dc81286c38e
62 Nc8f57c96b6344a0090f7d9eec95ad4e4
63 Ne3815ce85fe14f05b41ac8f0c5f813cf
64 schema:sameAs https://app.dimensions.ai/details/publication/pub.1111775678
65 https://doi.org/10.1186/s42490-019-0003-2
66 schema:sdDatePublished 2019-04-11T08:58
67 schema:sdLicense https://scigraph.springernature.com/explorer/license/
68 schema:sdPublisher N623ecec6023c43b39d47864961146e4c
69 schema:url https://link.springer.com/10.1186%2Fs42490-019-0003-2
70 sgo:license sg:explorer/license/
71 sgo:sdDataset articles
72 rdf:type schema:ScholarlyArticle
73 N0020a08eb65d4459af7489cc4d3b8e3e rdf:first N07a32379aa6d47b5a2cdcc001be3fc4d
74 rdf:rest N9c8c8cb8415c463993e7cd0efd45ba4f
75 N07a32379aa6d47b5a2cdcc001be3fc4d schema:affiliation https://www.grid.ac/institutes/grid.414699.7
76 schema:familyName Vigueras-Guillén
77 schema:givenName Juan P.
78 rdf:type schema:Person
79 N2ab12d46cf2a480d80df5fa69778ce37 schema:affiliation https://www.grid.ac/institutes/grid.5292.c
80 schema:familyName van Vliet
81 schema:givenName Lucas J.
82 rdf:type schema:Person
83 N39d7eb3182c243f79bf98929876e5cde schema:issueNumber 1
84 rdf:type schema:PublicationIssue
85 N39dd804aa9c041a7ad24c2c21b910d13 rdf:first Nf528173b6b704975bd9f7859de39c664
86 rdf:rest Nb0cf6d2ee733481795536cdd72f5fb79
87 N479d75044fa14aa5882f8bde585cb309 schema:affiliation https://www.grid.ac/institutes/grid.414699.7
88 schema:familyName Vermeer
89 schema:givenName Koenraad A.
90 rdf:type schema:Person
91 N4841d01270cc45ac98aa4dc81286c38e schema:name doi
92 schema:value 10.1186/s42490-019-0003-2
93 rdf:type schema:PropertyValue
94 N623ecec6023c43b39d47864961146e4c schema:name Springer Nature - SN SciGraph project
95 rdf:type schema:Organization
96 N6ff8866025234d90be7e24e3380b878e rdf:first N2ab12d46cf2a480d80df5fa69778ce37
97 rdf:rest rdf:nil
98 N9c8c8cb8415c463993e7cd0efd45ba4f rdf:first N9d6e58488aad43f69c34c9e07a25dd61
99 rdf:rest Nb659f417a8af45b5b2f6708a727d9eb9
100 N9d6e58488aad43f69c34c9e07a25dd61 schema:affiliation https://www.grid.ac/institutes/grid.5292.c
101 schema:familyName Sari
102 schema:givenName Busra
103 rdf:type schema:Person
104 Na40bb2e9849445cbb07028aa54ba199d schema:volumeNumber 1
105 rdf:type schema:PublicationVolume
106 Nb0cf6d2ee733481795536cdd72f5fb79 rdf:first N479d75044fa14aa5882f8bde585cb309
107 rdf:rest N6ff8866025234d90be7e24e3380b878e
108 Nb659f417a8af45b5b2f6708a727d9eb9 rdf:first Nb736723b4c09484ca49affdab177fffa
109 rdf:rest Nd66cc6b7044c4e4d9a8028f08413c2c8
110 Nb736723b4c09484ca49affdab177fffa schema:affiliation https://www.grid.ac/institutes/grid.5292.c
111 schema:familyName Goes
112 schema:givenName Stanley F.
113 rdf:type schema:Person
114 Nc8f57c96b6344a0090f7d9eec95ad4e4 schema:name readcube_id
115 schema:value 8b1d748b6c91ab573167c17cfc2db048731c72c1bd3b6219142f2221889cc711
116 rdf:type schema:PropertyValue
117 Nd66cc6b7044c4e4d9a8028f08413c2c8 rdf:first Nec61bf5eef1f4ab3920d1becfedc8ced
118 rdf:rest N39dd804aa9c041a7ad24c2c21b910d13
119 Ne3815ce85fe14f05b41ac8f0c5f813cf schema:name dimensions_id
120 schema:value pub.1111775678
121 rdf:type schema:PropertyValue
122 Nec61bf5eef1f4ab3920d1becfedc8ced schema:affiliation https://www.grid.ac/institutes/grid.414699.7
123 schema:familyName Lemij
124 schema:givenName Hans G.
125 rdf:type schema:Person
126 Nf528173b6b704975bd9f7859de39c664 schema:affiliation https://www.grid.ac/institutes/grid.414699.7
127 schema:familyName van Rooij
128 schema:givenName Jeroen
129 rdf:type schema:Person
130 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
131 schema:name Information and Computing Sciences
132 rdf:type schema:DefinedTerm
133 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
134 schema:name Artificial Intelligence and Image Processing
135 rdf:type schema:DefinedTerm
136 sg:journal.1336331 schema:issn 2524-4426
137 schema:name BMC Biomedical Engineering
138 rdf:type schema:Periodical
139 sg:pub.10.1007/978-1-4419-0300-6 schema:sameAs https://app.dimensions.ai/details/publication/pub.1027055606
140 https://doi.org/10.1007/978-1-4419-0300-6
141 rdf:type schema:CreativeWork
142 sg:pub.10.1007/978-3-319-23814-2_14 schema:sameAs https://app.dimensions.ai/details/publication/pub.1005182543
143 https://doi.org/10.1007/978-3-319-23814-2_14
144 rdf:type schema:CreativeWork
145 sg:pub.10.1007/978-3-319-24574-4_28 schema:sameAs https://app.dimensions.ai/details/publication/pub.1017774818
146 https://doi.org/10.1007/978-3-319-24574-4_28
147 rdf:type schema:CreativeWork
148 sg:pub.10.1007/978-3-319-93000-8_72 schema:sameAs https://app.dimensions.ai/details/publication/pub.1104412424
149 https://doi.org/10.1007/978-3-319-93000-8_72
150 rdf:type schema:CreativeWork
151 sg:pub.10.1038/sj.eye.6700559 schema:sameAs https://app.dimensions.ai/details/publication/pub.1004950511
152 https://doi.org/10.1038/sj.eye.6700559
153 rdf:type schema:CreativeWork
154 sg:pub.10.1186/s12880-015-0054-3 schema:sameAs https://app.dimensions.ai/details/publication/pub.1001794110
155 https://doi.org/10.1186/s12880-015-0054-3
156 rdf:type schema:CreativeWork
157 https://app.dimensions.ai/details/publication/pub.1027055606 schema:CreativeWork
158 https://app.dimensions.ai/details/publication/pub.1078511631 schema:CreativeWork
159 https://app.dimensions.ai/details/publication/pub.1083061562 schema:CreativeWork
160 https://doi.org/10.1002/jmor.1051740104 schema:sameAs https://app.dimensions.ai/details/publication/pub.1007940510
161 rdf:type schema:CreativeWork
162 https://doi.org/10.1016/j.ajo.2009.04.016 schema:sameAs https://app.dimensions.ai/details/publication/pub.1023619767
163 rdf:type schema:CreativeWork
164 https://doi.org/10.1016/j.ajo.2011.10.024 schema:sameAs https://app.dimensions.ai/details/publication/pub.1043433263
165 rdf:type schema:CreativeWork
166 https://doi.org/10.1016/j.artmed.2018.04.004 schema:sameAs https://app.dimensions.ai/details/publication/pub.1103422892
167 rdf:type schema:CreativeWork
168 https://doi.org/10.1016/j.cmpb.2015.09.003 schema:sameAs https://app.dimensions.ai/details/publication/pub.1025329845
169 rdf:type schema:CreativeWork
170 https://doi.org/10.1016/j.cmpb.2018.03.015 schema:sameAs https://app.dimensions.ai/details/publication/pub.1101697241
171 rdf:type schema:CreativeWork
172 https://doi.org/10.1016/j.compmedimag.2016.07.010 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006504630
173 rdf:type schema:CreativeWork
174 https://doi.org/10.1016/s0010-4825(99)00010-4 schema:sameAs https://app.dimensions.ai/details/publication/pub.1015179935
175 rdf:type schema:CreativeWork
176 https://doi.org/10.1016/s0031-3203(00)00074-1 schema:sameAs https://app.dimensions.ai/details/publication/pub.1017163976
177 rdf:type schema:CreativeWork
178 https://doi.org/10.1016/s0734-189x(87)80186-x schema:sameAs https://app.dimensions.ai/details/publication/pub.1019291816
179 rdf:type schema:CreativeWork
180 https://doi.org/10.1097/00003226-200308000-00005 schema:sameAs https://app.dimensions.ai/details/publication/pub.1009308127
181 rdf:type schema:CreativeWork
182 https://doi.org/10.1097/01.ico.0000151505.03824.6c schema:sameAs https://app.dimensions.ai/details/publication/pub.1039134662
183 rdf:type schema:CreativeWork
184 https://doi.org/10.1097/icl.0000000000000362 schema:sameAs https://app.dimensions.ai/details/publication/pub.1084194696
185 rdf:type schema:CreativeWork
186 https://doi.org/10.1097/ico.0000000000000385 schema:sameAs https://app.dimensions.ai/details/publication/pub.1039021041
187 rdf:type schema:CreativeWork
188 https://doi.org/10.1097/ico.0000000000000908 schema:sameAs https://app.dimensions.ai/details/publication/pub.1038391544
189 rdf:type schema:CreativeWork
190 https://doi.org/10.1097/ico.0b013e31802be629 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006805600
191 rdf:type schema:CreativeWork
192 https://doi.org/10.1097/ico.0b013e31825de8fa schema:sameAs https://app.dimensions.ai/details/publication/pub.1045412299
193 rdf:type schema:CreativeWork
194 https://doi.org/10.1109/cbms.1992.245000 schema:sameAs https://app.dimensions.ai/details/publication/pub.1086365009
195 rdf:type schema:CreativeWork
196 https://doi.org/10.1109/icdar.2003.1227801 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094714779
197 rdf:type schema:CreativeWork
198 https://doi.org/10.1109/icpr.1994.576361 schema:sameAs https://app.dimensions.ai/details/publication/pub.1094517408
199 rdf:type schema:CreativeWork
200 https://doi.org/10.1109/iembs.2007.4353724 schema:sameAs https://app.dimensions.ai/details/publication/pub.1077517956
201 rdf:type schema:CreativeWork
202 https://doi.org/10.1109/tfuzz.2016.2574915 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061607167
203 rdf:type schema:CreativeWork
204 https://doi.org/10.1109/tmi.2018.2841910 schema:sameAs https://app.dimensions.ai/details/publication/pub.1104318661
205 rdf:type schema:CreativeWork
206 https://doi.org/10.1109/tpami.2016.2572683 schema:sameAs https://app.dimensions.ai/details/publication/pub.1061745111
207 rdf:type schema:CreativeWork
208 https://doi.org/10.1117/12.2254342 schema:sameAs https://app.dimensions.ai/details/publication/pub.1084714918
209 rdf:type schema:CreativeWork
210 https://doi.org/10.1117/12.2264430 schema:sameAs https://app.dimensions.ai/details/publication/pub.1085407403
211 rdf:type schema:CreativeWork
212 https://doi.org/10.1117/12.60644 schema:sameAs https://app.dimensions.ai/details/publication/pub.1050290921
213 rdf:type schema:CreativeWork
214 https://doi.org/10.1136/bjo.2009.166561 schema:sameAs https://app.dimensions.ai/details/publication/pub.1043676105
215 rdf:type schema:CreativeWork
216 https://doi.org/10.1145/3065386 schema:sameAs https://app.dimensions.ai/details/publication/pub.1085642448
217 rdf:type schema:CreativeWork
218 https://doi.org/10.1162/neco.1989.1.4.541 schema:sameAs https://app.dimensions.ai/details/publication/pub.1008345178
219 rdf:type schema:CreativeWork
220 https://doi.org/10.15439/2017f54 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095851541
221 rdf:type schema:CreativeWork
222 https://doi.org/10.2307/1932409 schema:sameAs https://app.dimensions.ai/details/publication/pub.1069656769
223 rdf:type schema:CreativeWork
224 https://doi.org/10.23915/distill.00003 schema:sameAs https://app.dimensions.ai/details/publication/pub.1087157001
225 rdf:type schema:CreativeWork
226 https://doi.org/10.3349/ymj.2011.52.2.322 schema:sameAs https://app.dimensions.ai/details/publication/pub.1033041955
227 rdf:type schema:CreativeWork
228 https://doi.org/10.3390/sym10030060 schema:sameAs https://app.dimensions.ai/details/publication/pub.1101350073
229 rdf:type schema:CreativeWork
230 https://doi.org/10.3928/15428877-20110812-04 schema:sameAs https://app.dimensions.ai/details/publication/pub.1071728620
231 rdf:type schema:CreativeWork
232 https://doi.org/10.5566/ias.v27.p53-61 schema:sameAs https://app.dimensions.ai/details/publication/pub.1072988760
233 rdf:type schema:CreativeWork
234 https://www.grid.ac/institutes/grid.414699.7 schema:alternateName Rotterdam Eye Hospital
235 schema:name Delft University of Technology, Dept. of Imaging Physics, Lorentzweg 1, 2628CJ, Delft, The Netherlands
236 Rotterdam Ophthalmic Institute, Schiedamse Vest 160, 3011BH, Rotterdam, The Netherlands
237 The Rotterdam Eye Hospital, Schiedamse Vest 180, 3011BH, Rotterdam, The Netherlands
238 rdf:type schema:Organization
239 https://www.grid.ac/institutes/grid.5292.c schema:alternateName Delft University of Technology
240 schema:name Delft University of Technology, Dept. of Imaging Physics, Lorentzweg 1, 2628CJ, Delft, The Netherlands
241 rdf:type schema:Organization
 




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


...