Learning Probabilistic Fusion of Multilabel Lesion Contours

Gal Cohen; Hayit Greenspan; Jacob Goldberger

doi:10.1109/isbi45749.2020.9098540

Learning Probabilistic Fusion of Multilabel Lesion Contours

Gal Cohen, Hayit Greenspan, Jacob Goldberger

Bar-Ilan University - The Alexander Kofkin Faculty of Engineering

Tel Aviv University

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

3 Scopus citations

Abstract

Supervised machine learning algorithms, especially in the medical domain, are affected by considerable ambiguity in expert markings, primarily in proximity to lesion contours. In this study we address the case where the experts opinion for those ambiguous areas is considered as a distribution over the possible values. We propose a novel method that modifies the experts' distributional opinion at ambiguous areas by fusing their markings based on their sensitivity and specificity. The algorithm can be applied at the end of any label fusion algorithm that can handle soft values. The algorithm was applied to obtain consensus from soft Multiple Sclerosis (MS) segmentation masks. Soft MS segmentations are constructed from manual binary delineations by including lesion surrounding voxels in the segmentation mask with a reduced confidence weight. The method was evaluated on the MICCAI 2016 challenge dataset, and outperformed previous methods.

Original language	English
Title of host publication	ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging
Publisher	IEEE Computer Society
Pages	767-771
Number of pages	5
ISBN (Electronic)	9781538693308
DOIs	https://doi.org/10.1109/isbi45749.2020.9098540
State	Published - Apr 2020
Event	17th IEEE International Symposium on Biomedical Imaging, ISBI 2020 - Iowa City, United States Duration: 3 Apr 2020 → 7 Apr 2020

Publication series

Name	Proceedings - International Symposium on Biomedical Imaging
Volume	2020-April
ISSN (Print)	1945-7928
ISSN (Electronic)	1945-8452

Conference

Conference	17th IEEE International Symposium on Biomedical Imaging, ISBI 2020
Country/Territory	United States
City	Iowa City
Period	3/04/20 → 7/04/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Funding

This research was supported by the Ministry of Science & Technology, Israel

Funders	Funder number
Ministry of science and technology, Israel

Keywords

multiple annotators
multiple sclerosis
segmentation
soft labels

Access to Document

10.1109/isbi45749.2020.9098540

Cite this

@inproceedings{3141a895595a471a82c3bdc9ab323283,

title = "Learning Probabilistic Fusion of Multilabel Lesion Contours",

abstract = "Supervised machine learning algorithms, especially in the medical domain, are affected by considerable ambiguity in expert markings, primarily in proximity to lesion contours. In this study we address the case where the experts opinion for those ambiguous areas is considered as a distribution over the possible values. We propose a novel method that modifies the experts' distributional opinion at ambiguous areas by fusing their markings based on their sensitivity and specificity. The algorithm can be applied at the end of any label fusion algorithm that can handle soft values. The algorithm was applied to obtain consensus from soft Multiple Sclerosis (MS) segmentation masks. Soft MS segmentations are constructed from manual binary delineations by including lesion surrounding voxels in the segmentation mask with a reduced confidence weight. The method was evaluated on the MICCAI 2016 challenge dataset, and outperformed previous methods.",

keywords = "multiple annotators, multiple sclerosis, segmentation, soft labels",

author = "Gal Cohen and Hayit Greenspan and Jacob Goldberger",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020 ; Conference date: 03-04-2020 Through 07-04-2020",

year = "2020",

month = apr,

doi = "10.1109/isbi45749.2020.9098540",

language = "אנגלית",

series = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE Computer Society",

pages = "767--771",

booktitle = "ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging",

address = "ארצות הברית",

}

Cohen, G, Greenspan, H & Goldberger, J 2020, Learning Probabilistic Fusion of Multilabel Lesion Contours. in ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging., 9098540, Proceedings - International Symposium on Biomedical Imaging, vol. 2020-April, IEEE Computer Society, pp. 767-771, 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020, Iowa City, United States, 3/04/20. https://doi.org/10.1109/isbi45749.2020.9098540

Learning Probabilistic Fusion of Multilabel Lesion Contours. / Cohen, Gal; Greenspan, Hayit; Goldberger, Jacob.
ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2020. p. 767-771 9098540 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2020-April).

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

TY - GEN

T1 - Learning Probabilistic Fusion of Multilabel Lesion Contours

AU - Cohen, Gal

AU - Greenspan, Hayit

AU - Goldberger, Jacob

PY - 2020/4

Y1 - 2020/4

N2 - Supervised machine learning algorithms, especially in the medical domain, are affected by considerable ambiguity in expert markings, primarily in proximity to lesion contours. In this study we address the case where the experts opinion for those ambiguous areas is considered as a distribution over the possible values. We propose a novel method that modifies the experts' distributional opinion at ambiguous areas by fusing their markings based on their sensitivity and specificity. The algorithm can be applied at the end of any label fusion algorithm that can handle soft values. The algorithm was applied to obtain consensus from soft Multiple Sclerosis (MS) segmentation masks. Soft MS segmentations are constructed from manual binary delineations by including lesion surrounding voxels in the segmentation mask with a reduced confidence weight. The method was evaluated on the MICCAI 2016 challenge dataset, and outperformed previous methods.

AB - Supervised machine learning algorithms, especially in the medical domain, are affected by considerable ambiguity in expert markings, primarily in proximity to lesion contours. In this study we address the case where the experts opinion for those ambiguous areas is considered as a distribution over the possible values. We propose a novel method that modifies the experts' distributional opinion at ambiguous areas by fusing their markings based on their sensitivity and specificity. The algorithm can be applied at the end of any label fusion algorithm that can handle soft values. The algorithm was applied to obtain consensus from soft Multiple Sclerosis (MS) segmentation masks. Soft MS segmentations are constructed from manual binary delineations by including lesion surrounding voxels in the segmentation mask with a reduced confidence weight. The method was evaluated on the MICCAI 2016 challenge dataset, and outperformed previous methods.

KW - multiple annotators

KW - multiple sclerosis

KW - segmentation

KW - soft labels

UR - http://www.scopus.com/inward/record.url?scp=85085866680&partnerID=8YFLogxK

U2 - 10.1109/isbi45749.2020.9098540

DO - 10.1109/isbi45749.2020.9098540

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85085866680

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 767

EP - 771

BT - ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging

PB - IEEE Computer Society

T2 - 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020

Y2 - 3 April 2020 through 7 April 2020

ER -

Learning Probabilistic Fusion of Multilabel Lesion Contours

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this