Page 181 - MEGIN Book Of Abstracts - 2023

Australia; Graeme Clark Institute for Biomedical Engineer- Keywords: Beamforming, Epilepsy surgery, Interictal
ing, The University of Melbourne, Parkville, VIC, Australia; spikes, MEG, Source localization
Department of Biomedical Engineering, The University of
Melbourne, Parkville, VIC, Australia; Department of Medicine, Clinical neurophysiology: official journal of the
The University of Melbourne, Fitzroy, VIC, Australia; Graeme International Federation of Clinical Neurophysiology
Clark Institute for Biomedical Engineering, The University of (2021), Vol. 132, No. 4 (33636608) (2 citations)
Melbourne, Parkville, VIC, Australia

OBJECTIVE Magnetoencephalography (MEG) kurtosis A computational biomarker of juvenile myoclonic
beamforming is an automated localization method epilepsy from resting-state MEG (2021)
for focal epilepsy. Visual examination of virtual sen-
sors, which are source activities reconstructed by Lopes, Marinho A; Krzemiński, Dominik; Hamandi,
beamforming, can improve performance but can be Khalid; Singh, Krish D; Masuda, Naoki; Terry, John R;
time-consuming for neurophysiologists. We propose Zhang, Jiaxiang
a framework to automate the method and evaluate its
effectiveness against surgical resections and outcomes. Cardiff University Brain Research Imaging Centre, School of
Psychology, Cardiff University, Cardiff CF24 4HQ, United King-
METHODS We retrospectively analyzed MEG record- dom. Electronic address: [email protected]; Cardiff University
ings of 13 epilepsy surgery patients who had one-year Brain Research Imaging Centre, School of Psychology, Cardiff
minimum post-operative follow-up. Kurtosis beam- University, Cardiff CF24 4HQ, United Kingdom; The Welsh
forming was applied and manual inspection was Epilepsy Unit, Department of Neurology, University Hospital
confined to morphological clusters. The region with of Wales, Cardiff CF14 4XW, United Kingdom; Department of
the Maximum Interictal Spike Frequency (MISF) was Mathematics, University at Buffalo, State University of New
validated against prospectively modelled sLORETA York, USA; Computational and Data-Enabled Science and
solutions and surgical resections linked to outcome. Engineering Program, University at Buffalo, State University
of New York, USA; EPSRC Centre for Predictive Modelling in
RESULTS Our approach localized spikes in 12 out of 13 Healthcare, University of Birmingham, Birmingham, United
patients. In eight patients with Engel I surgical out- Kingdom; Centre for Systems Modelling and Quantitative
comes, beamforming MISF regions were concordant Biomedicine, University of Birmingham, Edgbaston, United
with surgical resection at overlap level for five patients Kingdom; Institute for Metabolism and Systems Research,
and at lobar level for three patients. The MISF regions University of Birmingham, Edgbaston, United Kingdom
localized to spike onset and propagation modelled by
sLORETA in two and six patients, respectively. OBJECTIVE For people with idiopathic generalized epi-
lepsy, functional networks derived from their resting-
CONCLUSIONS Automated beamforming using MEG state scalp electrophysiological recordings have shown
can predict postoperative seizure freedom at the lobar an inherent higher propensity to generate seizures
level but tends to localize propagated MEG spikes. than those from healthy controls when assessed using
the concept of brain network ictogenicity (BNI). Herein
SIGNIFICANCE MEG beamforming may contribute to we tested whether the BNI framework is applicable to
non-invasive procedures to predict surgical outcome resting-state magnetoencephalography (MEG) from
for patients with drug-refractory focal epilepsy. people with juvenile myoclonic epilepsy (JME).

METHODS The BNI framework consists in deriving a
functional network from apparently normal brain activ-
ity, placing a mathematical model of ictogenicity into
the network and then computing how often such net-

 ontents  Index 160
C