Forskning ved Københavns Universitet - Københavns Universitet

Forside

Detecting emphysema with multiple instance learning

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

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.

OriginalsprogEngelsk
Titel2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
ForlagIEEE
Publikationsdato23 maj 2018
Sider510-513
ISBN (Elektronisk)9781538636367
DOI
StatusUdgivet - 23 maj 2018
Begivenhed15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 - Washington, USA
Varighed: 4 apr. 20187 apr. 2018

Konference

Konference15th IEEE International Symposium on Biomedical Imaging, ISBI 2018
LandUSA
ByWashington
Periode04/04/201807/04/2018

ID: 199968017