Mean Field Network Based Graph Refinement with Application to Airway Tree Extraction View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2018-09-26

AUTHORS

Raghavendra Selvan , Max Welling , Jesper H. Pedersen , Jens Petersen , Marleen de Bruijne

ABSTRACT

We present tree extraction in 3D images as a graph refinement task, of obtaining a subgraph from an over-complete input graph. To this end, we formulate an approximate Bayesian inference framework on undirected graphs using mean field approximation (MFA). Mean field networks are used for inference based on the interpretation that iterations of MFA can be seen as feed-forward operations in a neural network. This allows us to learn the model parameters from training data using back-propagation algorithm. We demonstrate usefulness of the model to extract airway trees from 3D chest CT data. We first obtain probability images using a voxel classifier that distinguishes airways from background and use Bayesian smoothing to model individual airway branches. This yields us joint Gaussian density estimates of position, orientation and scale as node features of the input graph. Performance of the method is compared with two methods: the first uses probability images from a trained voxel classifier with region growing, which is similar to one of the best performing methods at EXACT’09 airway challenge, and the second method is based on Bayesian smoothing on these probability images. Using centerline distance as error measure the presented method shows significant improvement compared to these two methods. More... »

PAGES

750-758

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-030-00934-2_83

DOI

http://dx.doi.org/10.1007/978-3-030-00934-2_83

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1107031537


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

JSON-LD is the canonical representation for SciGraph data.

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

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0104", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Statistics", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Department of Computer Science, University of Copenhagen, Copenhagen, Denmark", 
          "id": "http://www.grid.ac/institutes/grid.5254.6", 
          "name": [
            "Department of Computer Science, University of Copenhagen, Copenhagen, Denmark"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Selvan", 
        "givenName": "Raghavendra", 
        "id": "sg:person.015077255170.84", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015077255170.84"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Canadian Institute for Advanced Research, Toronto, Canada", 
          "id": "http://www.grid.ac/institutes/grid.440050.5", 
          "name": [
            "Informatics Institute, University of Amsterdam, Amsterdam, Netherlands", 
            "Canadian Institute for Advanced Research, Toronto, Canada"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Welling", 
        "givenName": "Max", 
        "id": "sg:person.01315203510.28", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01315203510.28"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Department of Cardio-Thoracic Surgery RT, University Hospital of Copenhagen, Copenhagen, Denmark", 
          "id": "http://www.grid.ac/institutes/grid.4973.9", 
          "name": [
            "Department of Cardio-Thoracic Surgery RT, University Hospital of Copenhagen, Copenhagen, Denmark"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Pedersen", 
        "givenName": "Jesper H.", 
        "id": "sg:person.07725433554.11", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07725433554.11"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Department of Computer Science, University of Copenhagen, Copenhagen, Denmark", 
          "id": "http://www.grid.ac/institutes/grid.5254.6", 
          "name": [
            "Department of Computer Science, University of Copenhagen, Copenhagen, Denmark"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Petersen", 
        "givenName": "Jens", 
        "id": "sg:person.0621164750.96", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0621164750.96"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Departments of Medical Informatics and Radiology, Erasmus Medical Center, Rotterdam, Netherlands", 
          "id": "http://www.grid.ac/institutes/grid.5645.2", 
          "name": [
            "Department of Computer Science, University of Copenhagen, Copenhagen, Denmark", 
            "Departments of Medical Informatics and Radiology, Erasmus Medical Center, Rotterdam, Netherlands"
          ], 
          "type": "Organization"
        }, 
        "familyName": "de Bruijne", 
        "givenName": "Marleen", 
        "id": "sg:person.013733372770.12", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013733372770.12"
        ], 
        "type": "Person"
      }
    ], 
    "datePublished": "2018-09-26", 
    "datePublishedReg": "2018-09-26", 
    "description": "We present tree extraction in 3D images as a graph refinement task, of obtaining a subgraph from an over-complete input graph. To this end, we formulate an approximate Bayesian inference framework on undirected graphs using mean field approximation (MFA). Mean field networks are used for inference based on the interpretation that iterations of MFA can be seen as feed-forward operations in a neural network. This allows us to learn the model parameters from training data using back-propagation algorithm. We demonstrate usefulness of the model to extract airway trees from 3D chest CT data. We first obtain probability images using a voxel classifier that distinguishes airways from background and use Bayesian smoothing to model individual airway branches. This yields us joint Gaussian density estimates of position, orientation and scale as node features of the input graph. Performance of the method is compared with two methods: the first uses probability images from a trained voxel classifier with region growing, which is similar to one of the best performing methods at EXACT\u201909 airway challenge, and the second method is based on Bayesian smoothing on these probability images. Using centerline distance as error measure the presented method shows significant improvement compared to these two methods.", 
    "editor": [
      {
        "familyName": "Frangi", 
        "givenName": "Alejandro F.", 
        "type": "Person"
      }, 
      {
        "familyName": "Schnabel", 
        "givenName": "Julia A.", 
        "type": "Person"
      }, 
      {
        "familyName": "Davatzikos", 
        "givenName": "Christos", 
        "type": "Person"
      }, 
      {
        "familyName": "Alberola-L\u00f3pez", 
        "givenName": "Carlos", 
        "type": "Person"
      }, 
      {
        "familyName": "Fichtinger", 
        "givenName": "Gabor", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-030-00934-2_83", 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-030-00933-5", 
        "978-3-030-00934-2"
      ], 
      "name": "Medical Image Computing and Computer Assisted Intervention \u2013 MICCAI 2018", 
      "type": "Book"
    }, 
    "keywords": [
      "mean-field approximation", 
      "probability images", 
      "voxel classifier", 
      "input graph", 
      "tree extraction", 
      "approximate Bayesian inference framework", 
      "mean-field network", 
      "Bayesian inference framework", 
      "back-propagation algorithm", 
      "graph refinement", 
      "chest CT data", 
      "node features", 
      "refinement task", 
      "neural network", 
      "field approximation", 
      "region growing", 
      "field network", 
      "undirected graph", 
      "inference framework", 
      "model parameters", 
      "classifier", 
      "error measures", 
      "graph", 
      "images", 
      "airway tree", 
      "airway branches", 
      "CT data", 
      "network", 
      "second method", 
      "centerline distance", 
      "presented method", 
      "density estimates", 
      "approximation", 
      "algorithm", 
      "iteration", 
      "extraction", 
      "Bayesian", 
      "task", 
      "subgraphs", 
      "framework", 
      "inference", 
      "method", 
      "significant improvement", 
      "data", 
      "applications", 
      "performance", 
      "challenges", 
      "parameters", 
      "estimates", 
      "trees", 
      "operation", 
      "features", 
      "model", 
      "usefulness", 
      "branches", 
      "growing", 
      "distance", 
      "refinement", 
      "improvement", 
      "interpretation", 
      "end", 
      "orientation", 
      "position", 
      "scale", 
      "background", 
      "measures", 
      "feed", 
      "airway challenge"
    ], 
    "name": "Mean Field Network Based Graph Refinement with Application to Airway Tree Extraction", 
    "pagination": "750-758", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1107031537"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-030-00934-2_83"
        ]
      }
    ], 
    "publisher": {
      "name": "Springer Nature", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-030-00934-2_83", 
      "https://app.dimensions.ai/details/publication/pub.1107031537"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2022-11-24T21:12", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20221124/entities/gbq_results/chapter/chapter_175.jsonl", 
    "type": "Chapter", 
    "url": "https://doi.org/10.1007/978-3-030-00934-2_83"
  }
]
 

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.10.1007/978-3-030-00934-2_83'

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

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-00934-2_83'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-00934-2_83'

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

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-030-00934-2_83'


 

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

194 TRIPLES      22 PREDICATES      94 URIs      85 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-030-00934-2_83 schema:about anzsrc-for:01
2 anzsrc-for:0104
3 anzsrc-for:08
4 anzsrc-for:0801
5 schema:author Na37811cab93d450e9d9b9b7ac67a02e6
6 schema:datePublished 2018-09-26
7 schema:datePublishedReg 2018-09-26
8 schema:description We present tree extraction in 3D images as a graph refinement task, of obtaining a subgraph from an over-complete input graph. To this end, we formulate an approximate Bayesian inference framework on undirected graphs using mean field approximation (MFA). Mean field networks are used for inference based on the interpretation that iterations of MFA can be seen as feed-forward operations in a neural network. This allows us to learn the model parameters from training data using back-propagation algorithm. We demonstrate usefulness of the model to extract airway trees from 3D chest CT data. We first obtain probability images using a voxel classifier that distinguishes airways from background and use Bayesian smoothing to model individual airway branches. This yields us joint Gaussian density estimates of position, orientation and scale as node features of the input graph. Performance of the method is compared with two methods: the first uses probability images from a trained voxel classifier with region growing, which is similar to one of the best performing methods at EXACT’09 airway challenge, and the second method is based on Bayesian smoothing on these probability images. Using centerline distance as error measure the presented method shows significant improvement compared to these two methods.
9 schema:editor N68117448fcae4e9781082bc1648ca436
10 schema:genre chapter
11 schema:isAccessibleForFree true
12 schema:isPartOf Nfb21d442d2d74090abde8189fa2ea726
13 schema:keywords Bayesian
14 Bayesian inference framework
15 CT data
16 airway branches
17 airway challenge
18 airway tree
19 algorithm
20 applications
21 approximate Bayesian inference framework
22 approximation
23 back-propagation algorithm
24 background
25 branches
26 centerline distance
27 challenges
28 chest CT data
29 classifier
30 data
31 density estimates
32 distance
33 end
34 error measures
35 estimates
36 extraction
37 features
38 feed
39 field approximation
40 field network
41 framework
42 graph
43 graph refinement
44 growing
45 images
46 improvement
47 inference
48 inference framework
49 input graph
50 interpretation
51 iteration
52 mean-field approximation
53 mean-field network
54 measures
55 method
56 model
57 model parameters
58 network
59 neural network
60 node features
61 operation
62 orientation
63 parameters
64 performance
65 position
66 presented method
67 probability images
68 refinement
69 refinement task
70 region growing
71 scale
72 second method
73 significant improvement
74 subgraphs
75 task
76 tree extraction
77 trees
78 undirected graph
79 usefulness
80 voxel classifier
81 schema:name Mean Field Network Based Graph Refinement with Application to Airway Tree Extraction
82 schema:pagination 750-758
83 schema:productId N64b1998dd4504924b57e4d0817440ecf
84 Nb240406fbabc4569a25120167e4b8cf1
85 schema:publisher N7d51010e29a04161a8518940913350a3
86 schema:sameAs https://app.dimensions.ai/details/publication/pub.1107031537
87 https://doi.org/10.1007/978-3-030-00934-2_83
88 schema:sdDatePublished 2022-11-24T21:12
89 schema:sdLicense https://scigraph.springernature.com/explorer/license/
90 schema:sdPublisher N3b6f79bef71b472c80a8e6608bac3bc5
91 schema:url https://doi.org/10.1007/978-3-030-00934-2_83
92 sgo:license sg:explorer/license/
93 sgo:sdDataset chapters
94 rdf:type schema:Chapter
95 N013d2c95877246c4908459955b12a34f schema:familyName Frangi
96 schema:givenName Alejandro F.
97 rdf:type schema:Person
98 N133dfc92f66043019d7a8aa11dc3e757 rdf:first Nc0f770fa20524ac6a76a4c01fd3adae7
99 rdf:rest N86b48e3d0e5f4db182e26f4759c91dad
100 N1ef59778b0f14545989c8c0ffa26ae5d rdf:first sg:person.0621164750.96
101 rdf:rest N3bb024f7fedc44ba85aaa327d5619ec6
102 N3b6f79bef71b472c80a8e6608bac3bc5 schema:name Springer Nature - SN SciGraph project
103 rdf:type schema:Organization
104 N3bb024f7fedc44ba85aaa327d5619ec6 rdf:first sg:person.013733372770.12
105 rdf:rest rdf:nil
106 N3d78b94486eb42ac8042f251578b9a45 rdf:first sg:person.01315203510.28
107 rdf:rest Necddc716e3d9467a895a92ca52c8b04c
108 N64b1998dd4504924b57e4d0817440ecf schema:name dimensions_id
109 schema:value pub.1107031537
110 rdf:type schema:PropertyValue
111 N68117448fcae4e9781082bc1648ca436 rdf:first N013d2c95877246c4908459955b12a34f
112 rdf:rest N83b088d483d54233889b0bdb38f2428b
113 N7d51010e29a04161a8518940913350a3 schema:name Springer Nature
114 rdf:type schema:Organisation
115 N83b088d483d54233889b0bdb38f2428b rdf:first Nac8ed5a69226451e83bc5197478bb6de
116 rdf:rest Nf08522e95f984cc8a1f4900251f1494d
117 N86b48e3d0e5f4db182e26f4759c91dad rdf:first Nae33d91a05e543819ac53798134af803
118 rdf:rest rdf:nil
119 Na37811cab93d450e9d9b9b7ac67a02e6 rdf:first sg:person.015077255170.84
120 rdf:rest N3d78b94486eb42ac8042f251578b9a45
121 Nac8ed5a69226451e83bc5197478bb6de schema:familyName Schnabel
122 schema:givenName Julia A.
123 rdf:type schema:Person
124 Nae33d91a05e543819ac53798134af803 schema:familyName Fichtinger
125 schema:givenName Gabor
126 rdf:type schema:Person
127 Nb240406fbabc4569a25120167e4b8cf1 schema:name doi
128 schema:value 10.1007/978-3-030-00934-2_83
129 rdf:type schema:PropertyValue
130 Nc0f770fa20524ac6a76a4c01fd3adae7 schema:familyName Alberola-López
131 schema:givenName Carlos
132 rdf:type schema:Person
133 Ncbdc9265de394cbaa00183005cea4ad2 schema:familyName Davatzikos
134 schema:givenName Christos
135 rdf:type schema:Person
136 Necddc716e3d9467a895a92ca52c8b04c rdf:first sg:person.07725433554.11
137 rdf:rest N1ef59778b0f14545989c8c0ffa26ae5d
138 Nf08522e95f984cc8a1f4900251f1494d rdf:first Ncbdc9265de394cbaa00183005cea4ad2
139 rdf:rest N133dfc92f66043019d7a8aa11dc3e757
140 Nfb21d442d2d74090abde8189fa2ea726 schema:isbn 978-3-030-00933-5
141 978-3-030-00934-2
142 schema:name Medical Image Computing and Computer Assisted Intervention – MICCAI 2018
143 rdf:type schema:Book
144 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
145 schema:name Mathematical Sciences
146 rdf:type schema:DefinedTerm
147 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
148 schema:name Statistics
149 rdf:type schema:DefinedTerm
150 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
151 schema:name Information and Computing Sciences
152 rdf:type schema:DefinedTerm
153 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
154 schema:name Artificial Intelligence and Image Processing
155 rdf:type schema:DefinedTerm
156 sg:person.01315203510.28 schema:affiliation grid-institutes:grid.440050.5
157 schema:familyName Welling
158 schema:givenName Max
159 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.01315203510.28
160 rdf:type schema:Person
161 sg:person.013733372770.12 schema:affiliation grid-institutes:grid.5645.2
162 schema:familyName de Bruijne
163 schema:givenName Marleen
164 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013733372770.12
165 rdf:type schema:Person
166 sg:person.015077255170.84 schema:affiliation grid-institutes:grid.5254.6
167 schema:familyName Selvan
168 schema:givenName Raghavendra
169 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015077255170.84
170 rdf:type schema:Person
171 sg:person.0621164750.96 schema:affiliation grid-institutes:grid.5254.6
172 schema:familyName Petersen
173 schema:givenName Jens
174 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.0621164750.96
175 rdf:type schema:Person
176 sg:person.07725433554.11 schema:affiliation grid-institutes:grid.4973.9
177 schema:familyName Pedersen
178 schema:givenName Jesper H.
179 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07725433554.11
180 rdf:type schema:Person
181 grid-institutes:grid.440050.5 schema:alternateName Canadian Institute for Advanced Research, Toronto, Canada
182 schema:name Canadian Institute for Advanced Research, Toronto, Canada
183 Informatics Institute, University of Amsterdam, Amsterdam, Netherlands
184 rdf:type schema:Organization
185 grid-institutes:grid.4973.9 schema:alternateName Department of Cardio-Thoracic Surgery RT, University Hospital of Copenhagen, Copenhagen, Denmark
186 schema:name Department of Cardio-Thoracic Surgery RT, University Hospital of Copenhagen, Copenhagen, Denmark
187 rdf:type schema:Organization
188 grid-institutes:grid.5254.6 schema:alternateName Department of Computer Science, University of Copenhagen, Copenhagen, Denmark
189 schema:name Department of Computer Science, University of Copenhagen, Copenhagen, Denmark
190 rdf:type schema:Organization
191 grid-institutes:grid.5645.2 schema:alternateName Departments of Medical Informatics and Radiology, Erasmus Medical Center, Rotterdam, Netherlands
192 schema:name Department of Computer Science, University of Copenhagen, Copenhagen, Denmark
193 Departments of Medical Informatics and Radiology, Erasmus Medical Center, Rotterdam, Netherlands
194 rdf:type schema:Organization
 




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


...