Page 195 - MEGIN Book Of Abstracts - 2023

Recurrence quantification analysis of dynamic brain The relationship between neuromagnetic activity
networks (2021) and cognitive function in benign childhood
epilepsy with centrotemporal spikes (2020)
Lopes, Marinho A; Zhang, Jiaxiang; Krzemiński,
Dominik; Hamandi, Khalid; Chen, Qi; Livi, Lorenzo; Li, Yihan; Sun, Yulei; Niu, Kai; Wang, Pengfei; Xiang, Jing;
Masuda, Naoki Chen, Qiqi; Hu, Zheng; Wang, Xiaoshan

Cardiff University Brain Research Imaging Centre, School of Department of Neurology, Nanjing Brain Hospital, Nanjing
Psychology, Cardiff University, Cardiff, UK; Center for Studies Medical University, Nanjing, Jiangsu 210029, China; MEG
of Psychological Application and School of Psychology, South Center, Division of Neurology, Cincinnati Children's Hospital
China Normal University, Guangzhou, China; Department of Medical Center, Cincinnati, OH 45220, USA; MEG Center,
Computer Science, College of Engineering, Mathematics and Nanjing Brain Hospital, Nanjing, Jiangsu 210029, China; De-
Physical Sciences, University of Exeter, Exeter, UK; Computa- partment of Neurology, Nanjing Children's Hospital, Nanjing,
tional and Data-Enabled Science and Engineering Program, Jiangsu 210029, China; Department of Neurology, Nanjing
University at Buffalo, State University of New York, Buffalo, NY, Brain Hospital, Nanjing Medical University, Nanjing, Jiangsu
USA 210029, China. Electronic address: [email protected]

ABSTRACT Evidence suggests that brain network PURPOSE Our aim was to explore the pathophysiologi-
dynamics are a key determinant of brain function and cal mechanism of cognitive function changes in early
dysfunction. Here we propose a new framework to untreated children with benign childhood epilepsy
assess the dynamics of brain networks based on recur- with centrotemporal spikes (BECTS).
rence analysis. Our framework uses recurrence plots
and recurrence quantification analysis to characterize METHODS Magnetoencephalography (MEG) was
dynamic networks. For resting-state magnetoencepha- performed in 33 children with BECTS and 18 healthy
lographic dynamic functional networks (dFNs), we have children. Wechsler Intelligence Scale for Children,
found that functional networks recur more quickly fourth edition (WISC-IV) was used to divide children
in people with epilepsy than in healthy controls. This with BECTS into two groups. Normal cognitive function
suggests that recurrence of dFNs may be used as a was defined as a full-scale intelligence quotient (FSIQ)
biomarker of epilepsy. For stereo electroencephalogra- of >80, while decreased cognitive function was defined
phy data, we have found that dFNs involved in epileptic as a FSIQ of <80. Accumulated source imaging was
seizures emerge before seizure onset, and recurrence used to evaluate the neuromagnetic source activity in
analysis allows us to detect seizures. We further observe multifrequency bands.
distinct dFNs before and after seizures, which may
inform neurostimulation strategies to prevent seizures. RESULTS Of the 33 patients with early untreated
Our framework can also be used for understanding BECTS, a total of 17 had a FSIQ of <80 and 16 had FSIQ
dFNs in healthy brain function and in other neurologi- of >80. The course of epilepsy and number of seizures
cal disorders besides epilepsy. in the FSIQ <80 group were higher than that in the FSIQ
>80 group. Our MEG results showed that in the 4-8 Hz
Keywords: MEG, epilepsy, functional network, stereo EEG frequency band, both patient groups had inactivation
of the posterior cingulate cortex (PCC) region com-
The European journal of neuroscience (2021), Vol. 53, No. pared with the healthy control group. In the 30-80 Hz
4 (32888203) (4 citations) frequency band, the FSIQ <80 group showed inactiva-
tion of the PCC region compared with both the healthy
control group and the FSIQ >80 group. In the 80-250 Hz
frequency band, the FSIQ <80 group had inactivated of
the medial frontal cortex (MFC) region compared with

 ontents  Index 174
C