Unraveling the Relationship Between Beta Bursts and Hemodynamic Response During Motor Tasks
Poster No:
2075
Submission Type:
Abstract Submission
Authors:
Siyu Long1, Georgios Mitsis2, Marie-Hélène Boudrias3
Institutions:
1Integrated Program in Neuroscience, McGill University, Montreal, Quebec, 2Department of Bioengineering, McGill University, Montreal, Quebec, 3McGill University, Montreal, Quebec
First Author:
Co-Author(s):
Introduction:
Brain oscillations can manifest across multiple frequency bands in resting state and task-related states (Buzsaki, 2004). They can be interpreted as transient bursts rather than rhythmically sustained oscillations (Feingold, 2015). There are considerable evidences for the role of beta-band burst in voluntary movements. For example, beta bursts in a time-limited window of the contralateral subthalamic nucleus could reduce the peak velocity of the voluntary movement (Torrecillos, 2018) and beta burst timing was proved to be a strong predictor of single trial voluntary movement performance (Little, 2019). In this case, beta bursts could be indicative electrophysiological biomarkers of motor tasks as they are associated with motor performance.
The concentration of deoxyhemoglobin and oxyhemoglobin changed due to the onset of neural activity in the motor tasks (Jasdzewski, 2003). Some studies have shown temporal concurrence between electrophysiological and hemodynamic response (HDR) signals' features through simultaneously recording electroencephalographic (EEG) and Functional Magnetic Resonance Imaging (fMRI) data (Hunyadi, 2019). However, limited knowledge exist relationship between transient events (beta bursts) in EEG signals and HDR in fMRI signals.
Here we aimed to estimate the hemodynamic response function (HRF) based on beta bursts and calculate the mean squared error (MSE) between the blood-oxygen-level-dependent (BOLD) signal and the estimated signal.
The concentration of deoxyhemoglobin and oxyhemoglobin changed due to the onset of neural activity in the motor tasks (Jasdzewski, 2003). Some studies have shown temporal concurrence between electrophysiological and hemodynamic response (HDR) signals' features through simultaneously recording electroencephalographic (EEG) and Functional Magnetic Resonance Imaging (fMRI) data (Hunyadi, 2019). However, limited knowledge exist relationship between transient events (beta bursts) in EEG signals and HDR in fMRI signals.
Here we aimed to estimate the hemodynamic response function (HRF) based on beta bursts and calculate the mean squared error (MSE) between the blood-oxygen-level-dependent (BOLD) signal and the estimated signal.
Methods:
Eleven participants (age range 20-29) took part in a hand grip task while EEG was continuously recorded during the fMRI scanning. Both EEG and fMRI data were preprocessed via a standard pipeline (Prokopiou, 2022).
EEG source space was reconstructed for each subject using an extension of the linearly constrained minimum variance (LCMV) beamformer (Van, 1997). The Desikan-Killiany-Tourville (DKT) Atlas was implemented for all the sources and combined sources from the same parcel using Principal Component Analysis (PCA). A realistic head model for each subject was obtained using the subject's individual cortical anatomy and precise electrode locations on the scalp. Lead fields were estimated using the symmetric boundary element method (BEM).
To extract beta bursts, a threshold was applied, which corresponded to the signal's envelope followed by a second threshold reflecting the duration of the fluctuations or bursts. the beta signal was extracted and the 75th percentile value of the absolute beta envelope was used as the threshold. The minimum burst duration was set as 100 ms to account for rapid fluctuations that may result in false bursts (Tinkhauser, 2017).
Spherical Laguerre basis function was used to estimate the Hemodynamic Response Function (HRF) between beta bursts and BOLD signal in each brain region (Leistedt, 2012).
EEG source space was reconstructed for each subject using an extension of the linearly constrained minimum variance (LCMV) beamformer (Van, 1997). The Desikan-Killiany-Tourville (DKT) Atlas was implemented for all the sources and combined sources from the same parcel using Principal Component Analysis (PCA). A realistic head model for each subject was obtained using the subject's individual cortical anatomy and precise electrode locations on the scalp. Lead fields were estimated using the symmetric boundary element method (BEM).
To extract beta bursts, a threshold was applied, which corresponded to the signal's envelope followed by a second threshold reflecting the duration of the fluctuations or bursts. the beta signal was extracted and the 75th percentile value of the absolute beta envelope was used as the threshold. The minimum burst duration was set as 100 ms to account for rapid fluctuations that may result in false bursts (Tinkhauser, 2017).
Spherical Laguerre basis function was used to estimate the Hemodynamic Response Function (HRF) between beta bursts and BOLD signal in each brain region (Leistedt, 2012).
Results:
We found the MSE was the lowest in the visual cortex followed by the precentral and postcentral regions, while the MSE in other regions was relatively high (no significant difference). Besides, the HRF peak was positive in precentral, postcentral, middle frontal, superior temporal, insula, inferior parietal and cuneus regions, while other regions, such as lateral occipital, middle temporal and superior frontal regions showed negative peaks (Fig.1). Group average HRF estimation obtained in three parcels where beta bursts predicted the BOLD signal best are shown in Fig.2. Representative BOLD signal predictions obtained from one subject for the left precentral cortex is also shown in the same Figure. It suggested that beta bursts could obtain reliable prediction of HRF as well as the BOLD signal.
Conclusions:
Beta bursts reflect the hemodynamic response, especially in the motor and visual regions. This study provides a deeper understanding of the mechanisms and relationships between hemodynamics and electrophysiology during the movement execution, indicating that beta bursts can be a biological indicator of hemodynamic dynamics.
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 2
Motor Behavior:
Motor Planning and Execution 1
Keywords:
Electroencephaolography (EEG)
FUNCTIONAL MRI
Modeling
Motor
Source Localization
1|2Indicates the priority used for review
Provide references using author date format
Feingold J, et al. Bursts of beta oscillation differentiate postperformance activity in the striatum and motor cortex of monkeys performing movement tasks[J]. Proceedings of the National Academy of Sciences, 2015, 112(44): 13687-13692.
Hunyadi A. The mechanism (s) of action of antioxidants: From scavenging reactive oxygen/nitrogen species to redox signaling and the generation of bioactive secondary metabolites[J]. Medicinal research reviews, 2019, 39(6): 2505-2533.
Jasdzewski G, et al. Differences in the hemodynamic response to event-related motor and visual paradigms as measured by near-infrared spectroscopy[J]. Neuroimage, 2003, 20(1): 479-488.
Leistedt B, et al. Exact wavelets on the ball[J]. IEEE Transactions on Signal Processing, 2012, 60(12): 6257-6269.
Little S, et al. Human motor cortical beta bursts relate to movement planning and response errors[J]. PLoS biology, 2019, 17(10): e3000479.
Prokopiou P C, et al. Modeling the hemodynamic response function using EEG-fMRI data during eyes-open resting-state conditions and motor task execution[J]. Brain Topography, 2022, 35(3): 302-321.
Tinkhauser G, et al. Beta burst dynamics in Parkinson’s disease OFF and ON dopaminergic medication[J]. Brain, 2017, 140(11): 2968-2981.
Torrecillos F, et al. Modulation of beta bursts in the subthalamic nucleus predicts motor performance[J]. Journal of neuroscience, 2018, 38(41): 8905-8917.
Van Veen B D, et al. Localization of brain electrical activity via linearly constrained minimum variance spatial filtering[J]. IEEE Transactions on biomedical engineering, 1997, 44(9): 867-880.