Fully convolutional network for liver segmentation and lesions detection

Avi Ben-Cohen; Idit Diamant; Eyal Klang; Michal Amitai; Hayit Greenspan

doi:10.1007/978-3-319-46976-8_9

Fully convolutional network for liver segmentation and lesions detection

Avi Ben-Cohen, Idit Diamant, Eyal Klang, Michal Amitai, Hayit Greenspan

Tel Aviv University

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

Abstract

In this work we explore a fully convolutional network (FCN) for the task of liver segmentation and liver metastases detection in computed tomography (CT) examinations. FCN has proven to be a very powerful tool for semantic segmentation. We explore the FCN performance on a relatively small dataset and compare it to patch based CNN and sparsity based classification schemes. Our data contains CT examinations from 20 patients with overall 68 lesions and 43 livers marked in one slice and 20 different patients with a full 3D liver segmentation. We ran 3-fold cross-validation and results indicate superiority of the FCN over all other methods tested. Using our fully automatic algorithm we achieved true positive rate of 0.86 and 0.6 false positive per case which are very promising and clinically relevant results.

Original language	English
Title of host publication	Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings
Editors	Zhi Lu, Vasileios Belagiannis, Joao Manuel R.S. Tavares, Jaime S. Cardoso, Andrew Bradley, Joao Paulo Papa, Jacinto C. Nascimento, Marco Loog, Julien Cornebise, Gustavo Carneiro, Diana Mateus, Loic Peter
Publisher	Springer Verlag
Pages	77-85
Number of pages	9
ISBN (Print)	9783319469751
DOIs	https://doi.org/10.1007/978-3-319-46976-8_9
State	Published - 2016
Event	1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference ... - Athens, Greece Duration: 21 Oct 2016 → 21 Oct 2016

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10008 LNCS

Conference

Conference	1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference ...
Country/Territory	Greece
City	Athens
Period	21/10/16 → 21/10/16

Keywords

CT
Deep learning
Detection
Liver lesions

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-319-46976-8_9

Cite this

Ben-Cohen, A., Diamant, I., Klang, E., Amitai, M., & Greenspan, H. (2016). Fully convolutional network for liver segmentation and lesions detection. In Z. Lu, V. Belagiannis, J. M. R. S. Tavares, J. S. Cardoso, A. Bradley, J. P. Papa, J. C. Nascimento, M. Loog, J. Cornebise, G. Carneiro, D. Mateus, & L. Peter (Eds.), Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings (pp. 77-85). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10008 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-46976-8_9

Ben-Cohen, Avi ; Diamant, Idit ; Klang, Eyal et al. / Fully convolutional network for liver segmentation and lesions detection. Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings. editor / Zhi Lu ; Vasileios Belagiannis ; Joao Manuel R.S. Tavares ; Jaime S. Cardoso ; Andrew Bradley ; Joao Paulo Papa ; Jacinto C. Nascimento ; Marco Loog ; Julien Cornebise ; Gustavo Carneiro ; Diana Mateus ; Loic Peter. Springer Verlag, 2016. pp. 77-85 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{8412e1abaf734bd981a346af8f00c365,

title = "Fully convolutional network for liver segmentation and lesions detection",

abstract = "In this work we explore a fully convolutional network (FCN) for the task of liver segmentation and liver metastases detection in computed tomography (CT) examinations. FCN has proven to be a very powerful tool for semantic segmentation. We explore the FCN performance on a relatively small dataset and compare it to patch based CNN and sparsity based classification schemes. Our data contains CT examinations from 20 patients with overall 68 lesions and 43 livers marked in one slice and 20 different patients with a full 3D liver segmentation. We ran 3-fold cross-validation and results indicate superiority of the FCN over all other methods tested. Using our fully automatic algorithm we achieved true positive rate of 0.86 and 0.6 false positive per case which are very promising and clinically relevant results.",

keywords = "CT, Deep learning, Detection, Liver lesions",

author = "Avi Ben-Cohen and Idit Diamant and Eyal Klang and Michal Amitai and Hayit Greenspan",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2016.; 1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference ... ; Conference date: 21-10-2016 Through 21-10-2016",

year = "2016",

doi = "10.1007/978-3-319-46976-8_9",

language = "الإنجليزيّة",

isbn = "9783319469751",

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

publisher = "Springer Verlag",

pages = "77--85",

editor = "Zhi Lu and Vasileios Belagiannis and Tavares, {Joao Manuel R.S.} and Cardoso, {Jaime S.} and Andrew Bradley and Papa, {Joao Paulo} and Nascimento, {Jacinto C.} and Marco Loog and Julien Cornebise and Gustavo Carneiro and Diana Mateus and Loic Peter",

booktitle = "Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings",

address = "ألمانيا",

}

Ben-Cohen, A, Diamant, I, Klang, E, Amitai, M & Greenspan, H 2016, Fully convolutional network for liver segmentation and lesions detection. in Z Lu, V Belagiannis, JMRS Tavares, JS Cardoso, A Bradley, JP Papa, JC Nascimento, M Loog, J Cornebise, G Carneiro, D Mateus & L Peter (eds), Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10008 LNCS, Springer Verlag, pp. 77-85, 1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference ..., Athens, Greece, 21/10/16. https://doi.org/10.1007/978-3-319-46976-8_9

Fully convolutional network for liver segmentation and lesions detection. / Ben-Cohen, Avi; Diamant, Idit; Klang, Eyal et al.
Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings. ed. / Zhi Lu; Vasileios Belagiannis; Joao Manuel R.S. Tavares; Jaime S. Cardoso; Andrew Bradley; Joao Paulo Papa; Jacinto C. Nascimento; Marco Loog; Julien Cornebise; Gustavo Carneiro; Diana Mateus; Loic Peter. Springer Verlag, 2016. p. 77-85 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10008 LNCS).

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

TY - GEN

T1 - Fully convolutional network for liver segmentation and lesions detection

AU - Ben-Cohen, Avi

AU - Diamant, Idit

AU - Klang, Eyal

AU - Amitai, Michal

AU - Greenspan, Hayit

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

PY - 2016

Y1 - 2016

N2 - In this work we explore a fully convolutional network (FCN) for the task of liver segmentation and liver metastases detection in computed tomography (CT) examinations. FCN has proven to be a very powerful tool for semantic segmentation. We explore the FCN performance on a relatively small dataset and compare it to patch based CNN and sparsity based classification schemes. Our data contains CT examinations from 20 patients with overall 68 lesions and 43 livers marked in one slice and 20 different patients with a full 3D liver segmentation. We ran 3-fold cross-validation and results indicate superiority of the FCN over all other methods tested. Using our fully automatic algorithm we achieved true positive rate of 0.86 and 0.6 false positive per case which are very promising and clinically relevant results.

AB - In this work we explore a fully convolutional network (FCN) for the task of liver segmentation and liver metastases detection in computed tomography (CT) examinations. FCN has proven to be a very powerful tool for semantic segmentation. We explore the FCN performance on a relatively small dataset and compare it to patch based CNN and sparsity based classification schemes. Our data contains CT examinations from 20 patients with overall 68 lesions and 43 livers marked in one slice and 20 different patients with a full 3D liver segmentation. We ran 3-fold cross-validation and results indicate superiority of the FCN over all other methods tested. Using our fully automatic algorithm we achieved true positive rate of 0.86 and 0.6 false positive per case which are very promising and clinically relevant results.

KW - CT

KW - Deep learning

KW - Detection

KW - Liver lesions

UR - http://www.scopus.com/inward/record.url?scp=84992504713&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-46976-8_9

DO - 10.1007/978-3-319-46976-8_9

M3 - منشور من مؤتمر

SN - 9783319469751

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

SP - 77

EP - 85

BT - Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings

A2 - Lu, Zhi

A2 - Belagiannis, Vasileios

A2 - Tavares, Joao Manuel R.S.

A2 - Cardoso, Jaime S.

A2 - Bradley, Andrew

A2 - Papa, Joao Paulo

A2 - Nascimento, Jacinto C.

A2 - Loog, Marco

A2 - Cornebise, Julien

A2 - Carneiro, Gustavo

A2 - Mateus, Diana

A2 - Peter, Loic

PB - Springer Verlag

T2 - 1st International Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016 and 2nd International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016 held in conjunction with 19th International Conference ...

Y2 - 21 October 2016 through 21 October 2016

ER -

Ben-Cohen A, Diamant I, Klang E, Amitai M, Greenspan H. Fully convolutional network for liver segmentation and lesions detection. In Lu Z, Belagiannis V, Tavares JMRS, Cardoso JS, Bradley A, Papa JP, Nascimento JC, Loog M, Cornebise J, Carneiro G, Mateus D, Peter L, editors, Deep Learning and Data Labeling for Medical Applications - 1st International Workshop, LABELS 2016, and 2nd International Workshop, DLMIA 2016 Held in Conjunction with MICCAI 2016, Proceedings. Springer Verlag. 2016. p. 77-85. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-46976-8_9

Fully convolutional network for liver segmentation and lesions detection

Abstract

Publication series

Conference

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this