U-Net: Convolutional Networks for Biomedical Image Segmentation View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2015-11-18

AUTHORS

Olaf Ronneberger , Philipp Fischer , Thomas Brox

ABSTRACT

There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net. More... »

PAGES

234-241

Book

TITLE

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015

ISBN

978-3-319-24573-7
978-3-319-24574-4

Patents

  • Method For Providing An Evaluating Means For At Least One Optical Application System Of A Microscopic Application Technology
  • Fast And Robust Friction Ridge Impression Minutiae Extraction Using Feed-Forward Convolutional Neural Network
  • Sampling Latent Variables To Generate Multiple Segmentations Of An Image
  • Medical Image Object Detection With Dense Feature Pyramid Network Architecture In Machine Learning
  • Method For Microscopic Assessment
  • Computer Method And Apparatus Making Screens Safe For Those With Photosensitivity
  • Training Data Generation For Artificial Intelligence-Based Sequencing
  • Detection Of Biological Cells And Tracing Of Cell Lineage
  • Artificial Intelligence-Based Sequencing
  • Modifying A Neural Network
  • System And Method For Single Image Object Density Estimation
  • Adaptive Specimen Image Acquisition Using An Artificial Neural Network
  • A System And Method For Single Image Object Density Estimation
  • Method For Determining Image Values In Marked Pixels Of At Least One Projection Image, X-Ray Device, Computer Program And Electronically Readable Data Carrier
  • Interactive Iterative Image Annotation
  • System And Method For Deriving High-Resolution Subsurface Reservoir Parameters
  • Method, Device And Computer Program For Creating A Neuronal Network
  • Modifying A Neural Network
  • Method For Identifying Bone Images
  • Learning Method And Learning Device For Improving Performance Of Cnn By Using Feature Upsampling Networks, And Testing Method And Testing Device Using The Same
  • Standardized Cloud Radiotherapy Planning Method, Storage Medium, And System
  • Method And Device For Generating Feature Maps By Using Feature Upsampling Networks
  • Artificial Intelligence-Based Generation Of Sequencing Metadata
  • Convolutional Neural Networks For Locating Objects Of Interest In Images Of Biological Samples
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/978-3-319-24574-4_28

    DOI

    http://dx.doi.org/10.1007/978-3-319-24574-4_28

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1017774818


    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/2008/08", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Information and Computing Sciences", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Artificial Intelligence and Image Processing", 
            "type": "DefinedTerm"
          }
        ], 
        "author": [
          {
            "affiliation": {
              "alternateName": "Computer Science Department and BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany", 
              "id": "http://www.grid.ac/institutes/grid.5963.9", 
              "name": [
                "Computer Science Department and BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Ronneberger", 
            "givenName": "Olaf", 
            "id": "sg:person.0625370723.52", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0625370723.52"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Computer Science Department and BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany", 
              "id": "http://www.grid.ac/institutes/grid.5963.9", 
              "name": [
                "Computer Science Department and BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Fischer", 
            "givenName": "Philipp", 
            "id": "sg:person.012106015125.15", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012106015125.15"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Computer Science Department and BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany", 
              "id": "http://www.grid.ac/institutes/grid.5963.9", 
              "name": [
                "Computer Science Department and BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Brox", 
            "givenName": "Thomas", 
            "id": "sg:person.012443225372.65", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012443225372.65"
            ], 
            "type": "Person"
          }
        ], 
        "datePublished": "2015-11-18", 
        "datePublishedReg": "2015-11-18", 
        "description": "There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net.", 
        "editor": [
          {
            "familyName": "Navab", 
            "givenName": "Nassir", 
            "type": "Person"
          }, 
          {
            "familyName": "Hornegger", 
            "givenName": "Joachim", 
            "type": "Person"
          }, 
          {
            "familyName": "Wells", 
            "givenName": "William M.", 
            "type": "Person"
          }, 
          {
            "familyName": "Frangi", 
            "givenName": "Alejandro F.", 
            "type": "Person"
          }
        ], 
        "genre": "chapter", 
        "id": "sg:pub.10.1007/978-3-319-24574-4_28", 
        "inLanguage": "en", 
        "isAccessibleForFree": true, 
        "isPartOf": {
          "isbn": [
            "978-3-319-24573-7", 
            "978-3-319-24574-4"
          ], 
          "name": "Medical Image Computing and Computer-Assisted Intervention \u2013 MICCAI 2015", 
          "type": "Book"
        }, 
        "keywords": [
          "biomedical image segmentation", 
          "deep network", 
          "convolutional network", 
          "data augmentation", 
          "ISBI challenge", 
          "contracting path", 
          "image segmentation", 
          "recent GPU", 
          "training samples", 
          "large margin", 
          "same network", 
          "segmentation", 
          "network", 
          "training strategy", 
          "prior best method", 
          "successful training", 
          "images", 
          "strong use", 
          "light microscopy images", 
          "GPU", 
          "full implementation", 
          "nets", 
          "architecture", 
          "path", 
          "precise localization", 
          "implementation", 
          "best method", 
          "stack", 
          "challenges", 
          "microscopy images", 
          "training", 
          "context", 
          "end", 
          "method", 
          "augmentation", 
          "localization", 
          "strategies", 
          "use", 
          "categories", 
          "neuronal structures", 
          "structure", 
          "Freiburg", 
          "margin", 
          "consent", 
          "samples", 
          "paper"
        ], 
        "name": "U-Net: Convolutional Networks for Biomedical Image Segmentation", 
        "pagination": "234-241", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1017774818"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/978-3-319-24574-4_28"
            ]
          }
        ], 
        "publisher": {
          "name": "Springer Nature", 
          "type": "Organisation"
        }, 
        "sameAs": [
          "https://doi.org/10.1007/978-3-319-24574-4_28", 
          "https://app.dimensions.ai/details/publication/pub.1017774818"
        ], 
        "sdDataset": "chapters", 
        "sdDatePublished": "2022-05-10T10:39", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-springernature-scigraph/baseset/20220509/entities/gbq_results/chapter/chapter_161.jsonl", 
        "type": "Chapter", 
        "url": "https://doi.org/10.1007/978-3-319-24574-4_28"
      }
    ]
     

    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.10.1007/978-3-319-24574-4_28'

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

    curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-24574-4_28'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-24574-4_28'

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

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-319-24574-4_28'


     

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

    135 TRIPLES      23 PREDICATES      71 URIs      64 LITERALS      7 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/978-3-319-24574-4_28 schema:about anzsrc-for:08
    2 anzsrc-for:0801
    3 schema:author Nbc16900b8d5e4493a8a46f2a42d19d65
    4 schema:datePublished 2015-11-18
    5 schema:datePublishedReg 2015-11-18
    6 schema:description There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net.
    7 schema:editor N3ad45aced6544361aecf5253e02b1be7
    8 schema:genre chapter
    9 schema:inLanguage en
    10 schema:isAccessibleForFree true
    11 schema:isPartOf N05c7fa12cdfd41b4889a1d6907f86808
    12 schema:keywords Freiburg
    13 GPU
    14 ISBI challenge
    15 architecture
    16 augmentation
    17 best method
    18 biomedical image segmentation
    19 categories
    20 challenges
    21 consent
    22 context
    23 contracting path
    24 convolutional network
    25 data augmentation
    26 deep network
    27 end
    28 full implementation
    29 image segmentation
    30 images
    31 implementation
    32 large margin
    33 light microscopy images
    34 localization
    35 margin
    36 method
    37 microscopy images
    38 nets
    39 network
    40 neuronal structures
    41 paper
    42 path
    43 precise localization
    44 prior best method
    45 recent GPU
    46 same network
    47 samples
    48 segmentation
    49 stack
    50 strategies
    51 strong use
    52 structure
    53 successful training
    54 training
    55 training samples
    56 training strategy
    57 use
    58 schema:name U-Net: Convolutional Networks for Biomedical Image Segmentation
    59 schema:pagination 234-241
    60 schema:productId N4ab5c1918b2d41818731bda40e3accb0
    61 N5995a5c0879f463c89bb97d8f7e330a1
    62 schema:publisher Nf2870d6a1b7340c9951ea77f8dac5374
    63 schema:sameAs https://app.dimensions.ai/details/publication/pub.1017774818
    64 https://doi.org/10.1007/978-3-319-24574-4_28
    65 schema:sdDatePublished 2022-05-10T10:39
    66 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    67 schema:sdPublisher N66d0888e6df14e2786b94ef84285294d
    68 schema:url https://doi.org/10.1007/978-3-319-24574-4_28
    69 sgo:license sg:explorer/license/
    70 sgo:sdDataset chapters
    71 rdf:type schema:Chapter
    72 N05c7fa12cdfd41b4889a1d6907f86808 schema:isbn 978-3-319-24573-7
    73 978-3-319-24574-4
    74 schema:name Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015
    75 rdf:type schema:Book
    76 N0df30cbc5d874beaa7b33abcccbae96b rdf:first Nb69330d16edb446ea891de795bd14579
    77 rdf:rest Nf82e0d7f59b34ed0a6ce9747088d9c2c
    78 N20d0a2774c934d18a7f41925cf69d398 schema:familyName Wells
    79 schema:givenName William M.
    80 rdf:type schema:Person
    81 N225266ba6c0b4b7ba2366f47ad3f3d0b rdf:first sg:person.012443225372.65
    82 rdf:rest rdf:nil
    83 N2f556b8a0e4f443abc4ac3cac5c6e1cc schema:familyName Navab
    84 schema:givenName Nassir
    85 rdf:type schema:Person
    86 N3ad45aced6544361aecf5253e02b1be7 rdf:first N2f556b8a0e4f443abc4ac3cac5c6e1cc
    87 rdf:rest N0df30cbc5d874beaa7b33abcccbae96b
    88 N4ab5c1918b2d41818731bda40e3accb0 schema:name dimensions_id
    89 schema:value pub.1017774818
    90 rdf:type schema:PropertyValue
    91 N5995a5c0879f463c89bb97d8f7e330a1 schema:name doi
    92 schema:value 10.1007/978-3-319-24574-4_28
    93 rdf:type schema:PropertyValue
    94 N5d1f05b25f6f4e7c94ae77060c303744 rdf:first sg:person.012106015125.15
    95 rdf:rest N225266ba6c0b4b7ba2366f47ad3f3d0b
    96 N66d0888e6df14e2786b94ef84285294d schema:name Springer Nature - SN SciGraph project
    97 rdf:type schema:Organization
    98 N84a2f84bbac84ecaba9ccb4b7e91ef91 schema:familyName Frangi
    99 schema:givenName Alejandro F.
    100 rdf:type schema:Person
    101 Nb0ffe1af4c7c45d78752fff13b033677 rdf:first N84a2f84bbac84ecaba9ccb4b7e91ef91
    102 rdf:rest rdf:nil
    103 Nb69330d16edb446ea891de795bd14579 schema:familyName Hornegger
    104 schema:givenName Joachim
    105 rdf:type schema:Person
    106 Nbc16900b8d5e4493a8a46f2a42d19d65 rdf:first sg:person.0625370723.52
    107 rdf:rest N5d1f05b25f6f4e7c94ae77060c303744
    108 Nf2870d6a1b7340c9951ea77f8dac5374 schema:name Springer Nature
    109 rdf:type schema:Organisation
    110 Nf82e0d7f59b34ed0a6ce9747088d9c2c rdf:first N20d0a2774c934d18a7f41925cf69d398
    111 rdf:rest Nb0ffe1af4c7c45d78752fff13b033677
    112 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    113 schema:name Information and Computing Sciences
    114 rdf:type schema:DefinedTerm
    115 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
    116 schema:name Artificial Intelligence and Image Processing
    117 rdf:type schema:DefinedTerm
    118 sg:person.012106015125.15 schema:affiliation grid-institutes:grid.5963.9
    119 schema:familyName Fischer
    120 schema:givenName Philipp
    121 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012106015125.15
    122 rdf:type schema:Person
    123 sg:person.012443225372.65 schema:affiliation grid-institutes:grid.5963.9
    124 schema:familyName Brox
    125 schema:givenName Thomas
    126 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012443225372.65
    127 rdf:type schema:Person
    128 sg:person.0625370723.52 schema:affiliation grid-institutes:grid.5963.9
    129 schema:familyName Ronneberger
    130 schema:givenName Olaf
    131 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0625370723.52
    132 rdf:type schema:Person
    133 grid-institutes:grid.5963.9 schema:alternateName Computer Science Department and BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany
    134 schema:name Computer Science Department and BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg, Germany
    135 rdf:type schema:Organization
     




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


    ...