Please use this identifier to cite or link to this item: https://doi.org/10.15480/882.3824
DC FieldValueLanguage
dc.contributor.authorHolstein, Lennart-
dc.contributor.authorRiedl, Katharina Alina-
dc.contributor.authorWissel, Tobias-
dc.contributor.authorBrunner, Fabian J.-
dc.contributor.authorSchaefers, Klaus-
dc.contributor.authorGraß, Michael-
dc.contributor.authorBlankenberg, Stefan-
dc.contributor.authorSeiffert, Moritz-
dc.contributor.authorSchlaefer, Alexander-
dc.date.accessioned2021-10-14T05:18:06Z-
dc.date.available2021-10-14T05:18:06Z-
dc.date.issued2021-08-01-
dc.identifier.citationCurrent Directions in Biomedical Engineering 7 (1): 20211123, 96-100 (2021-08-01)de_DE
dc.identifier.issn2364-5504de_DE
dc.identifier.urihttp://hdl.handle.net/11420/10493-
dc.description.abstractKnowing the shape of vascular calcifications is crucial for appropriate planning and conductance of percutaneous coronary interventions. The clinical workflow can therefore benefit from automatic segmentation of calcified plaques in intravascular ultrasound (IVUS) images. To solve segmentation problems with convolutional neural networks (CNNs), large datasets are usually required. However, datasets are often rather small in the medical domain. Hence, developing and investigating methods for increasing CNN performance on small datasets can help on the way towards clinically relevant results. We compared two state-of-the-art CNN architectures for segmentation, U-Net and DeepLabV3, and investigated how incorporating auxiliary image data with vessel wall and lumen annotations improves the calcium segmentation performance by using these either for pretraining or multi-task training. DeepLabV3 outperforms U-Net with up to 6.3 % by means of the Dice coefficient and 36.5 % by means of the average Hausdorff distance. Using auxiliary data improves the segmentation performance in both cases, whereas the multi-task approach outperforms the pre-training approach. The improvements of the multi-task approach in contrast to not using auxiliary data at all is 5.7 % for the Dice coefficient and 42.9 % for the average Hausdorff distance. Automatic segmentation of calcified plaques in IVUS images is a demanding task due to their relatively small size compared to the image dimensions and due to visual ambiguities with other image structures. We showed that this problem can generally be tackled by CNNs. Furthermore, we were able to improve the performance by a multi-task learning approach with auxiliary segmentation data.en
dc.language.isoende_DE
dc.publisherDe Gruyterde_DE
dc.relation.ispartofCurrent directions in biomedical engineeringde_DE
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/de_DE
dc.subjectConvolutional neural networkde_DE
dc.subjectCoronary arteryde_DE
dc.subjectMulti-task learningde_DE
dc.subjectSmall datasetde_DE
dc.subjectVesselde_DE
dc.subject.ddc600: Technikde_DE
dc.subject.ddc610: Medizinde_DE
dc.titleDeep learning for calcium segmentation in intravascular ultrasound imagesde_DE
dc.typeArticlede_DE
dc.identifier.doi10.15480/882.3824-
dc.type.diniarticle-
dcterms.DCMITypeText-
tuhh.identifier.urnurn:nbn:de:gbv:830-882.0147344-
tuhh.oai.showtruede_DE
tuhh.abstract.englishKnowing the shape of vascular calcifications is crucial for appropriate planning and conductance of percutaneous coronary interventions. The clinical workflow can therefore benefit from automatic segmentation of calcified plaques in intravascular ultrasound (IVUS) images. To solve segmentation problems with convolutional neural networks (CNNs), large datasets are usually required. However, datasets are often rather small in the medical domain. Hence, developing and investigating methods for increasing CNN performance on small datasets can help on the way towards clinically relevant results. We compared two state-of-the-art CNN architectures for segmentation, U-Net and DeepLabV3, and investigated how incorporating auxiliary image data with vessel wall and lumen annotations improves the calcium segmentation performance by using these either for pretraining or multi-task training. DeepLabV3 outperforms U-Net with up to 6.3 % by means of the Dice coefficient and 36.5 % by means of the average Hausdorff distance. Using auxiliary data improves the segmentation performance in both cases, whereas the multi-task approach outperforms the pre-training approach. The improvements of the multi-task approach in contrast to not using auxiliary data at all is 5.7 % for the Dice coefficient and 42.9 % for the average Hausdorff distance. Automatic segmentation of calcified plaques in IVUS images is a demanding task due to their relatively small size compared to the image dimensions and due to visual ambiguities with other image structures. We showed that this problem can generally be tackled by CNNs. Furthermore, we were able to improve the performance by a multi-task learning approach with auxiliary segmentation data.de_DE
tuhh.publisher.doi10.1515/cdbme-2021-1021-
tuhh.publication.instituteMedizintechnische und Intelligente Systeme E-1de_DE
tuhh.identifier.doi10.15480/882.3824-
tuhh.type.opus(wissenschaftlicher) Artikel-
dc.type.driverarticle-
dc.type.casraiJournal Article-
tuhh.container.issue1de_DE
tuhh.container.volume7de_DE
tuhh.container.startpage96de_DE
tuhh.container.endpage100de_DE
dc.relation.projectMALEKA: Maschinelle Lernverfahren für die kardiovaskuläre Bildgebung auf der Grundlage des Programms für Innovation (PROFI) - Modul PROFI Transfer Plusde_DE
dc.rights.nationallicensefalsede_DE
dc.identifier.scopus2-s2.0-85114431339de_DE
tuhh.container.articlenumber20211123de_DE
local.status.inpressfalsede_DE
local.type.versionpublishedVersionde_DE
item.grantfulltextopen-
item.languageiso639-1en-
item.creatorOrcidHolstein, Lennart-
item.creatorOrcidRiedl, Katharina Alina-
item.creatorOrcidWissel, Tobias-
item.creatorOrcidBrunner, Fabian J.-
item.creatorOrcidSchaefers, Klaus-
item.creatorOrcidGraß, Michael-
item.creatorOrcidBlankenberg, Stefan-
item.creatorOrcidSeiffert, Moritz-
item.creatorOrcidSchlaefer, Alexander-
item.cerifentitytypePublications-
item.openairetypeArticle-
item.creatorGNDHolstein, Lennart-
item.creatorGNDRiedl, Katharina Alina-
item.creatorGNDWissel, Tobias-
item.creatorGNDBrunner, Fabian J.-
item.creatorGNDSchaefers, Klaus-
item.creatorGNDGraß, Michael-
item.creatorGNDBlankenberg, Stefan-
item.creatorGNDSeiffert, Moritz-
item.creatorGNDSchlaefer, Alexander-
item.mappedtypeArticle-
item.openairecristypehttp://purl.org/coar/resource_type/c_6501-
item.fulltextWith Fulltext-
crisitem.author.deptMedizintechnische und Intelligente Systeme E-1-
crisitem.author.deptMedizintechnische und Intelligente Systeme E-1-
crisitem.author.orcid0000-0003-0610-0347-
crisitem.author.parentorgStudiendekanat Elektrotechnik, Informatik und Mathematik-
crisitem.author.parentorgStudiendekanat Elektrotechnik, Informatik und Mathematik-
crisitem.project.funderHamburgische Investitions- und Förderbank-
crisitem.project.funderrorid00012xz55-
Appears in Collections:Publications with fulltext
Files in This Item:
File Description SizeFormat
10.1515_cdbme-2021-1021.pdfVerlags-PDF1,37 MBAdobe PDFView/Open
Thumbnail
Show simple item record

Page view(s)

54
Last Week
0
Last month
checked on Nov 29, 2021

Download(s)

14
checked on Nov 29, 2021

Google ScholarTM

Check

Note about this record

Cite this record

Export

This item is licensed under a Creative Commons License Creative Commons