Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN

Maayan Frid-Adar; Idit Diamant; Eyal Klang; Michal Amitai; Jacob Goldberger; Hayit Greenspan

doi:10.1007/978-3-319-67434-6_15

Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN

Maayan Frid-Adar, Idit Diamant, Eyal Klang, Michal Amitai, Jacob Goldberger, Hayit Greenspan

Tel Aviv University, Bar-Ilan University

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

Abstract

Automatic detection of liver lesions in CT images poses a great challenge for researchers. In this work we present a deep learning approach that models explicitly the variability within the non-lesion class, based on prior knowledge of the data, to support an automated lesion detection system. A multi-class convolutional neural network (CNN) is proposed to categorize input image patches into sub-categories of boundary and interior patches, the decisions of which are fused to reach a binary lesion vs non-lesion decision. For validation of our system, we use CT images of 132 livers and 498 lesions. Our approach shows highly improved detection results that outperform the state-of-the-art fully convolutional network. Automated computerized tools, as shown in this work, have the potential in the future to support the radiologists towards improved detection.

Original language	English
Title of host publication	Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings
Editors	Yiqiang Zhan, Wenjia Bai, Guorong Wu, Pierrick Coupe, Brent C. Munsell, Gerard Sanroma
Publisher	Springer Verlag
Pages	129-137
Number of pages	9
ISBN (Print)	9783319674339
DOIs	https://doi.org/10.1007/978-3-319-67434-6_15
State	Published - 2017
Event	3rd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2017 held in conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 - Quebec City, Canada Duration: 14 Sep 2017 → 14 Sep 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10530 LNCS

Conference

Conference	3rd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2017 held in conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017
Country/Territory	Canada
City	Quebec City
Period	14/09/17 → 14/09/17

Keywords

Computer-aided detection
Convolutional neural network
Detection
Liver lesion
Patch-based system

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-319-67434-6_15

Cite this

Frid-Adar, M., Diamant, I., Klang, E., Amitai, M., Goldberger, J., & Greenspan, H. (2017). Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN. In Y. Zhan, W. Bai, G. Wu, P. Coupe, B. C. Munsell, & G. Sanroma (Eds.), Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings (pp. 129-137). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10530 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-67434-6_15

Frid-Adar, Maayan ; Diamant, Idit ; Klang, Eyal et al. / Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN. Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings. editor / Yiqiang Zhan ; Wenjia Bai ; Guorong Wu ; Pierrick Coupe ; Brent C. Munsell ; Gerard Sanroma. Springer Verlag, 2017. pp. 129-137 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{462595334b634888ade648083eca4039,

title = "Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN",

abstract = "Automatic detection of liver lesions in CT images poses a great challenge for researchers. In this work we present a deep learning approach that models explicitly the variability within the non-lesion class, based on prior knowledge of the data, to support an automated lesion detection system. A multi-class convolutional neural network (CNN) is proposed to categorize input image patches into sub-categories of boundary and interior patches, the decisions of which are fused to reach a binary lesion vs non-lesion decision. For validation of our system, we use CT images of 132 livers and 498 lesions. Our approach shows highly improved detection results that outperform the state-of-the-art fully convolutional network. Automated computerized tools, as shown in this work, have the potential in the future to support the radiologists towards improved detection.",

keywords = "Computer-aided detection, Convolutional neural network, Detection, Liver lesion, Patch-based system",

author = "Maayan Frid-Adar and Idit Diamant and Eyal Klang and Michal Amitai and Jacob Goldberger and Hayit Greenspan",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 3rd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2017 held in conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017 ; Conference date: 14-09-2017 Through 14-09-2017",

year = "2017",

doi = "10.1007/978-3-319-67434-6\_15",

language = "الإنجليزيّة",

isbn = "9783319674339",

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

publisher = "Springer Verlag",

pages = "129--137",

editor = "Yiqiang Zhan and Wenjia Bai and Guorong Wu and Pierrick Coupe and Munsell, \{Brent C.\} and Gerard Sanroma",

booktitle = "Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings",

address = "ألمانيا",

}

Frid-Adar, M, Diamant, I, Klang, E, Amitai, M, Goldberger, J & Greenspan, H 2017, Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN. in Y Zhan, W Bai, G Wu, P Coupe, BC Munsell & G Sanroma (eds), Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10530 LNCS, Springer Verlag, pp. 129-137, 3rd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2017 held in conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017, Quebec City, Canada, 14/09/17. https://doi.org/10.1007/978-3-319-67434-6_15

Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN. / Frid-Adar, Maayan; Diamant, Idit; Klang, Eyal et al.
Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings. ed. / Yiqiang Zhan; Wenjia Bai; Guorong Wu; Pierrick Coupe; Brent C. Munsell; Gerard Sanroma. Springer Verlag, 2017. p. 129-137 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10530 LNCS).

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

TY - GEN

T1 - Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN

AU - Frid-Adar, Maayan

AU - Diamant, Idit

AU - Klang, Eyal

AU - Amitai, Michal

AU - Goldberger, Jacob

AU - Greenspan, Hayit

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - Automatic detection of liver lesions in CT images poses a great challenge for researchers. In this work we present a deep learning approach that models explicitly the variability within the non-lesion class, based on prior knowledge of the data, to support an automated lesion detection system. A multi-class convolutional neural network (CNN) is proposed to categorize input image patches into sub-categories of boundary and interior patches, the decisions of which are fused to reach a binary lesion vs non-lesion decision. For validation of our system, we use CT images of 132 livers and 498 lesions. Our approach shows highly improved detection results that outperform the state-of-the-art fully convolutional network. Automated computerized tools, as shown in this work, have the potential in the future to support the radiologists towards improved detection.

AB - Automatic detection of liver lesions in CT images poses a great challenge for researchers. In this work we present a deep learning approach that models explicitly the variability within the non-lesion class, based on prior knowledge of the data, to support an automated lesion detection system. A multi-class convolutional neural network (CNN) is proposed to categorize input image patches into sub-categories of boundary and interior patches, the decisions of which are fused to reach a binary lesion vs non-lesion decision. For validation of our system, we use CT images of 132 livers and 498 lesions. Our approach shows highly improved detection results that outperform the state-of-the-art fully convolutional network. Automated computerized tools, as shown in this work, have the potential in the future to support the radiologists towards improved detection.

KW - Computer-aided detection

KW - Convolutional neural network

KW - Detection

KW - Liver lesion

KW - Patch-based system

UR - http://www.scopus.com/inward/record.url?scp=85029425396&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-67434-6_15

DO - 10.1007/978-3-319-67434-6_15

M3 - منشور من مؤتمر

SN - 9783319674339

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

SP - 129

EP - 137

BT - Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings

A2 - Zhan, Yiqiang

A2 - Bai, Wenjia

A2 - Wu, Guorong

A2 - Coupe, Pierrick

A2 - Munsell, Brent C.

A2 - Sanroma, Gerard

PB - Springer Verlag

T2 - 3rd International Workshop on Patch-Based Techniques in Medical Imaging, Patch-MI 2017 held in conjunction with 20th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2017

Y2 - 14 September 2017 through 14 September 2017

ER -

Frid-Adar M, Diamant I, Klang E, Amitai M, Goldberger J , Greenspan H. Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN. In Zhan Y, Bai W, Wu G, Coupe P, Munsell BC, Sanroma G, editors, Patch-Based Techniques in Medical Imaging - 3rd International Workshop, Patch-MI 2017 Held in Conjunction with MICCAI 2017, Proceedings. Springer Verlag. 2017. p. 129-137. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-67434-6_15

Modeling the intra-class variability for liver lesion detection using a multi-class patch-based CNN

Abstract

Publication series

Conference

Keywords

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this