Vis enkel innførsel

dc.contributor.authorHashemi, Ali
dc.contributor.authorCai, Chang
dc.contributor.authorKutyniok, Gitta Astrid Hildegard
dc.contributor.authorMüller, Klaus R.
dc.contributor.authorNagarajan, Srikantan S.
dc.contributor.authorHaufe, Stefan
dc.date.accessioned2022-03-02T12:19:30Z
dc.date.available2022-03-02T12:19:30Z
dc.date.issued2021-10-01
dc.description.abstractMethods for electro- or magnetoencephalography (EEG/MEG) based brain source imaging (BSI) using sparse Bayesian learning (SBL) have been demonstrated to achieve excellent performance in situations with low numbers of distinct active sources, such as event-related designs. This paper extends the theory and practice of SBL in three important ways. First, we reformulate three existing SBL algorithms under the <i>majorization-minimization</i> (MM) framework. This unification perspective not only provides a useful theoretical framework for comparing different algorithms in terms of their convergence behavior, but also provides a principled recipe for constructing novel algorithms with specific properties by designing appropriate bounds of the Bayesian marginal likelihood function. Second, building on the MM principle, we propose a novel method called <i>Low</i>SNR-BSI that achieves favorable source reconstruction performance in low signal-to-noise-ratio (SNR) settings. Third, precise knowledge of the noise level is a crucial requirement for accurate source reconstruction. Here we present a novel principled technique to accurately learn the noise variance from the data either jointly within the source reconstruction procedure or using one of two proposed cross-validation strategies. Empirically, we could show that the monotonous convergence behavior predicted from MM theory is confirmed in numerical experiments. Using simulations, we further demonstrate the advantage of LowSNR-BSI over conventional SBL in low-SNR regimes, and the advantage of learned noise levels over estimates derived from baseline data. To demonstrate the usefulness of our novel approach, we show neurophysiologically plausible source reconstructions on averaged auditory evoked potential data.en_US
dc.description.sponsorshipEUen_US
dc.identifier.citationHashemi, Cai, Kutyniok, Müller, Nagarajan, Haufe. Unification of sparse Bayesian learning algorithms for electromagnetic brain imaging with the majorization minimization framework. NeuroImage. 2021;239en_US
dc.identifier.cristinIDFRIDAID 2000221
dc.identifier.doi10.1016/j.neuroimage.2021.118309
dc.identifier.issn1053-8119
dc.identifier.issn1095-9572
dc.identifier.urihttps://hdl.handle.net/10037/24222
dc.language.isoengen_US
dc.publisherElsevieren_US
dc.relation.journalNeuroImage
dc.relation.projectIDinfo:eu-repo/grantAgreement/ERC/H2020/758985/Germany/Advancing the non-invasive assessment of brain communication in neurological disease//en_US
dc.rights.accessRightsopenAccessen_US
dc.rights.holderCopyright 2021 The Author(s)en_US
dc.titleUnification of sparse Bayesian learning algorithms for electromagnetic brain imaging with the majorization minimization frameworken_US
dc.type.versionpublishedVersionen_US
dc.typeJournal articleen_US
dc.typeTidsskriftartikkelen_US
dc.typePeer revieweden_US


Tilhørende fil(er)

Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel