Detecting emphysema with multiple instance learning

Silas Nyboe Orting; Jens Petersen; Laura H. Thomsen; Mathilde M.W. Wille; Marleen De Bruijne

doi:10.1109/ISBI.2018.8363627

Detecting emphysema with multiple instance learning

Silas Nyboe Orting, Jens Petersen, Laura H. Thomsen, Mathilde M.W. Wille, Marleen De Bruijne

Department of Computer Science

4 Citations (Scopus)

Abstract

Emphysema is part of chronic obstructive pulmonary disease, a leading cause of mortality worldwide. Visual assessment of emphysema presence is useful for identifying subjects at risk and for research into disease development. We train a machine learning method to predict emphysema from visually assessed expert labels. We use a multiple instance learning approach to predict both scan-level and region-level emphysema presence. We evaluate performance on 600 low-dose CT scans from the Danish Lung Cancer Screening Study and achieve an AUC of 0.82 for scan-level prediction and AUCs between 0.76 and 0.88 for region-level prediction.

Original language	English
Title of host publication	2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
Publisher	IEEE
Publication date	23 May 2018
Pages	510-513
ISBN (Electronic)	9781538636367
DOIs	https://doi.org/10.1109/ISBI.2018.8363627
Publication status	Published - 23 May 2018
Event	15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 - Washington, United States Duration: 4 Apr 2018 → 7 Apr 2018

Conference

Conference	15th IEEE International Symposium on Biomedical Imaging, ISBI 2018
Country/Territory	United States
City	Washington
Period	04/04/2018 → 07/04/2018

Keywords

Emphysema
Multiple Instance Learning
Weak supervision

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ISBI.2018.8363627

Cite this

@inproceedings{7f0542b4d6e64f6f919386f94cd21f8b,

title = "Detecting emphysema with multiple instance learning",

abstract = "Emphysema is part of chronic obstructive pulmonary disease, a leading cause of mortality worldwide. Visual assessment of emphysema presence is useful for identifying subjects at risk and for research into disease development. We train a machine learning method to predict emphysema from visually assessed expert labels. We use a multiple instance learning approach to predict both scan-level and region-level emphysema presence. We evaluate performance on 600 low-dose CT scans from the Danish Lung Cancer Screening Study and achieve an AUC of 0.82 for scan-level prediction and AUCs between 0.76 and 0.88 for region-level prediction.",

keywords = "Emphysema, Multiple Instance Learning, Weak supervision",

author = "Orting, {Silas Nyboe} and Jens Petersen and Thomsen, {Laura H.} and Wille, {Mathilde M.W.} and {De Bruijne}, Marleen",

year = "2018",

month = may,

day = "23",

doi = "10.1109/ISBI.2018.8363627",

language = "English",

pages = "510--513",

booktitle = "2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018",

publisher = "IEEE",

note = "15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 ; Conference date: 04-04-2018 Through 07-04-2018",

}

TY - GEN

T1 - Detecting emphysema with multiple instance learning

AU - Orting, Silas Nyboe

AU - Petersen, Jens

AU - Thomsen, Laura H.

AU - Wille, Mathilde M.W.

AU - De Bruijne, Marleen

PY - 2018/5/23

Y1 - 2018/5/23

N2 - Emphysema is part of chronic obstructive pulmonary disease, a leading cause of mortality worldwide. Visual assessment of emphysema presence is useful for identifying subjects at risk and for research into disease development. We train a machine learning method to predict emphysema from visually assessed expert labels. We use a multiple instance learning approach to predict both scan-level and region-level emphysema presence. We evaluate performance on 600 low-dose CT scans from the Danish Lung Cancer Screening Study and achieve an AUC of 0.82 for scan-level prediction and AUCs between 0.76 and 0.88 for region-level prediction.

AB - Emphysema is part of chronic obstructive pulmonary disease, a leading cause of mortality worldwide. Visual assessment of emphysema presence is useful for identifying subjects at risk and for research into disease development. We train a machine learning method to predict emphysema from visually assessed expert labels. We use a multiple instance learning approach to predict both scan-level and region-level emphysema presence. We evaluate performance on 600 low-dose CT scans from the Danish Lung Cancer Screening Study and achieve an AUC of 0.82 for scan-level prediction and AUCs between 0.76 and 0.88 for region-level prediction.

KW - Emphysema

KW - Multiple Instance Learning

KW - Weak supervision

UR - http://www.scopus.com/inward/record.url?scp=85048089312&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2018.8363627

DO - 10.1109/ISBI.2018.8363627

M3 - Article in proceedings

AN - SCOPUS:85048089312

SP - 510

EP - 513

BT - 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018

PB - IEEE

T2 - 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018

Y2 - 4 April 2018 through 7 April 2018

ER -

Detecting emphysema with multiple instance learning

Abstract

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this