Show simple item record

dc.contributor.authorTomasetti, Luca
dc.contributor.authorKhanmohammadi, Mahdieh
dc.contributor.authorEngan, Kjersti
dc.contributor.authorHøllesli, Liv Jorunn
dc.contributor.authorKurz, Kathinka Dæhli
dc.identifier.citationTomasetti, L., Khanmohammadi, M., Engan, K., Høllesli, L.J., Kurz, K.D. (2022) Multi-input segmentation of damaged brain in acute ischemic stroke patients using slow fusion with skip connection. Proceedings of the Northern Lights Deep Learning Workshop 22, 3.en_US
dc.description.abstractTime is a fundamental factor during stroke treatments. A fast, automatic approach that segmentsthe ischemic regions helps treatment decisions. In clinical use today, a set of color-coded parametric maps generated from computed tomography perfusion (CTP) images are investigated manually to decide a treatment plan. We propose an automatic method based on a neural network using a set of parametric maps to segment the two ischemic regions (core and penumbra) in patients affected by acute ischemic stroke. Our model is based on a convolution-deconvolution bottleneck structure with multi-input and slow fusion. A loss function based on the focal Tversky index addresses the data imbalance issue. The proposed architecture demonstrates effective performance and results comparable to the ground truth annotated by neuroradiologists. A Dice coefficient of 0.81 for penumbra and 0.52 for core over the large vessel occlusion test set is achieved. The full implementation is available at:
dc.publisherSeptentrio Academic Publishingen_US
dc.rightsNavngivelse 4.0 Internasjonal*
dc.subjectischemic strokeen_US
dc.subjectimage segmentationen_US
dc.titleMulti-input segmentation of damaged brain in acute ischemic stroke patients using slow fusion with skip connectionen_US
dc.typePeer revieweden_US
dc.typeJournal articleen_US
dc.rights.holder© 2022 Luca Tomasetti, Mahdieh Khanmohammadi, Kjersti Engan, Liv Jorunn Høllesli, Kathinka Dæhli Kurzen_US
dc.subject.nsiVDP::Medisinske Fag: 700::Klinisk medisinske fag: 750::Nevrologi: 752en_US
dc.subject.nsiVDP::Teknologi: 500::Medisinsk teknologi: 620en_US
dc.source.journalProceedings of the Northern Lights Deep Learning Workshopen_US

Files in this item


This item appears in the following Collection(s)

Show simple item record

Navngivelse 4.0 Internasjonal
Except where otherwise noted, this item's license is described as Navngivelse 4.0 Internasjonal