Creating a large-scale silver corpus from multiple algorithmic segmentations

Markus Krenn; Matthias Dorfer; Oscar Alfonso Jiménez del Toro; Henning Müller; Bjoern Menze; Marc André Weber; Allan Hanbury; Georg Langs

doi:10.1007/978-3-319-42016-5_10

Creating a large-scale silver corpus from multiple algorithmic segmentations

Markus Krenn, Matthias Dorfer, Oscar Alfonso Jiménez del Toro, Henning Müller, Bjoern Menze, Marc André Weber, Allan Hanbury, Georg Langs

Default

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

Currently, increasingly large medical imaging data sets become available for research and are analysed by a range of algorithms segmenting anatomical structures automatically and interactively. While they provide segmentations on a much larger scale than possible to achieve with expert annotators, they are typically less accurate than experts. We present and compare approaches to estimate segmentations on large imaging data sets based on a small number of expert annotated examples, and algorithmic segmentations on a much larger data set. Results demonstrate that combining algorithmic segmentations is reliably outperforming the average individual algorithm. Furthermore, injecting organ specific reliability assessments of algorithms based on expert annotations improves accuracy compared to standard label fusion algorithms. The proposed methods are particularly relevant in putting the results of large image analysis algorithm benchmarks to long-term use.

Original language	English
Title of host publication	Medical Computer Vision
Subtitle of host publication	Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers
Editors	Michael Kelm, Henning Müller, Bjoern Menze, Shaoting Zhang, Dimitris Metaxas, Georg Langs, Albert Montillo, Weidong Cai
Publisher	Springer Verlag
Pages	103-115
Number of pages	13
ISBN (Print)	9783319420158
DOIs	https://doi.org/10.1007/978-3-319-42016-5_10
State	Published - 2016
Event	International Workshop on Medical Image Computing for Computer Assisted Intervention, 2015 MICCAI - Germany, Germany Duration: 9 Oct 2015 → 9 Oct 2015

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9601 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	International Workshop on Medical Image Computing for Computer Assisted Intervention, 2015 MICCAI
Country/Territory	Germany
City	Germany
Period	9/10/15 → 9/10/15

Keywords

Label fusion
Segmentation
Silver corpus

Access to Document

10.1007/978-3-319-42016-5_10

Cite this

Krenn, M., Dorfer, M., Jiménez del Toro, O. A., Müller, H., Menze, B., Weber, M. A., Hanbury, A., & Langs, G. (2016). Creating a large-scale silver corpus from multiple algorithmic segmentations. In M. Kelm, H. Müller, B. Menze, S. Zhang, D. Metaxas, G. Langs, A. Montillo, & W. Cai (Eds.), Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers (pp. 103-115). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9601 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-42016-5_10

Krenn, Markus ; Dorfer, Matthias ; Jiménez del Toro, Oscar Alfonso et al. / Creating a large-scale silver corpus from multiple algorithmic segmentations. Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers. editor / Michael Kelm ; Henning Müller ; Bjoern Menze ; Shaoting Zhang ; Dimitris Metaxas ; Georg Langs ; Albert Montillo ; Weidong Cai. Springer Verlag, 2016. pp. 103-115 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ce0fd6d1e8404095bb3cfcbffc8691ef,

title = "Creating a large-scale silver corpus from multiple algorithmic segmentations",

abstract = "Currently, increasingly large medical imaging data sets become available for research and are analysed by a range of algorithms segmenting anatomical structures automatically and interactively. While they provide segmentations on a much larger scale than possible to achieve with expert annotators, they are typically less accurate than experts. We present and compare approaches to estimate segmentations on large imaging data sets based on a small number of expert annotated examples, and algorithmic segmentations on a much larger data set. Results demonstrate that combining algorithmic segmentations is reliably outperforming the average individual algorithm. Furthermore, injecting organ specific reliability assessments of algorithms based on expert annotations improves accuracy compared to standard label fusion algorithms. The proposed methods are particularly relevant in putting the results of large image analysis algorithm benchmarks to long-term use.",

keywords = "Label fusion, Segmentation, Silver corpus",

author = "Markus Krenn and Matthias Dorfer and {Jim{\'e}nez del Toro}, {Oscar Alfonso} and Henning M{\"u}ller and Bjoern Menze and Weber, {Marc Andr{\'e}} and Allan Hanbury and Georg Langs",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2016.; International Workshop on Medical Image Computing for Computer Assisted Intervention, 2015 MICCAI ; Conference date: 09-10-2015 Through 09-10-2015",

year = "2016",

doi = "10.1007/978-3-319-42016-5_10",

language = "English",

isbn = "9783319420158",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "103--115",

editor = "Michael Kelm and Henning M{\"u}ller and Bjoern Menze and Shaoting Zhang and Dimitris Metaxas and Georg Langs and Albert Montillo and Weidong Cai",

booktitle = "Medical Computer Vision",

}

Krenn, M, Dorfer, M, Jiménez del Toro, OA, Müller, H, Menze, B, Weber, MA, Hanbury, A & Langs, G 2016, Creating a large-scale silver corpus from multiple algorithmic segmentations. in M Kelm, H Müller, B Menze, S Zhang, D Metaxas, G Langs, A Montillo & W Cai (eds), Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9601 LNCS, Springer Verlag, pp. 103-115, International Workshop on Medical Image Computing for Computer Assisted Intervention, 2015 MICCAI, Germany, Germany, 9/10/15. https://doi.org/10.1007/978-3-319-42016-5_10

Creating a large-scale silver corpus from multiple algorithmic segmentations. / Krenn, Markus; Dorfer, Matthias; Jiménez del Toro, Oscar Alfonso et al.
Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers. ed. / Michael Kelm; Henning Müller; Bjoern Menze; Shaoting Zhang; Dimitris Metaxas; Georg Langs; Albert Montillo; Weidong Cai. Springer Verlag, 2016. p. 103-115 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9601 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Creating a large-scale silver corpus from multiple algorithmic segmentations

AU - Krenn, Markus

AU - Dorfer, Matthias

AU - Jiménez del Toro, Oscar Alfonso

AU - Müller, Henning

AU - Menze, Bjoern

AU - Weber, Marc André

AU - Hanbury, Allan

AU - Langs, Georg

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2016.

PY - 2016

Y1 - 2016

N2 - Currently, increasingly large medical imaging data sets become available for research and are analysed by a range of algorithms segmenting anatomical structures automatically and interactively. While they provide segmentations on a much larger scale than possible to achieve with expert annotators, they are typically less accurate than experts. We present and compare approaches to estimate segmentations on large imaging data sets based on a small number of expert annotated examples, and algorithmic segmentations on a much larger data set. Results demonstrate that combining algorithmic segmentations is reliably outperforming the average individual algorithm. Furthermore, injecting organ specific reliability assessments of algorithms based on expert annotations improves accuracy compared to standard label fusion algorithms. The proposed methods are particularly relevant in putting the results of large image analysis algorithm benchmarks to long-term use.

AB - Currently, increasingly large medical imaging data sets become available for research and are analysed by a range of algorithms segmenting anatomical structures automatically and interactively. While they provide segmentations on a much larger scale than possible to achieve with expert annotators, they are typically less accurate than experts. We present and compare approaches to estimate segmentations on large imaging data sets based on a small number of expert annotated examples, and algorithmic segmentations on a much larger data set. Results demonstrate that combining algorithmic segmentations is reliably outperforming the average individual algorithm. Furthermore, injecting organ specific reliability assessments of algorithms based on expert annotations improves accuracy compared to standard label fusion algorithms. The proposed methods are particularly relevant in putting the results of large image analysis algorithm benchmarks to long-term use.

KW - Label fusion

KW - Segmentation

KW - Silver corpus

UR - http://www.scopus.com/inward/record.url?scp=84981341579&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-42016-5_10

DO - 10.1007/978-3-319-42016-5_10

M3 - Conference contribution

AN - SCOPUS:84981341579

SN - 9783319420158

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 103

EP - 115

BT - Medical Computer Vision

A2 - Kelm, Michael

A2 - Müller, Henning

A2 - Menze, Bjoern

A2 - Zhang, Shaoting

A2 - Metaxas, Dimitris

A2 - Langs, Georg

A2 - Montillo, Albert

A2 - Cai, Weidong

PB - Springer Verlag

T2 - International Workshop on Medical Image Computing for Computer Assisted Intervention, 2015 MICCAI

Y2 - 9 October 2015 through 9 October 2015

ER -

Krenn M, Dorfer M, Jiménez del Toro OA, Müller H, Menze B, Weber MA et al. Creating a large-scale silver corpus from multiple algorithmic segmentations. In Kelm M, Müller H, Menze B, Zhang S, Metaxas D, Langs G, Montillo A, Cai W, editors, Medical Computer Vision: Algorithms for Big Data - International Workshop, MCV 2015 and Held in Conjunction with MICCAI 2015, Revised Selected Papers. Springer Verlag. 2016. p. 103-115. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-42016-5_10

Creating a large-scale silver corpus from multiple algorithmic segmentations

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this