Characterization of Hand Clenching in Human Sensorimotor Cortex Using High-, and Ultra-High Frequency Band Modulations of Electrocorticogram

Abstract

Functional mapping of eloquent cortex before the resection of a tumor is a critical procedure for optimizing survival and quality of life. In order to locate the hand area of the motor cortex in two patients with low-grade gliomas (LGG), we recorded electrocorticogram (ECoG) from a 113 channel hybrid high-density grid (64 large contacts with diameter of 2.7 mm and 49 small contacts with diameter of 1 mm) while they executed hand clenching movements. We investigated the spatio-spectral characteristics of the neural oscillatory activity and observed that, in both patients, the hand movements were consistently associated with a wide spread power decrease in the low frequency band (LFB: 8–32 Hz) and a more localized power increase in the high frequency band (HFB: 60–280 Hz) within the sensorimotor region. Importantly, we observed significant power increase in the ultra-high frequency band (UFB: 300–800 Hz) during hand movements of both patients within a restricted cortical region close to the central sulcus, and the motor cortical “hand knob.” Among all frequency bands we studied, the UFB modulations were closest to the central sulcus and direct cortical stimulation (DCS) positive site. Both HFB and UFB modulations exhibited different timing characteristics at different locations. Power increase in HFB and UFB starting before movement onset was observed mostly at the anterior part of the activated cortical region. In addition, the spatial patterns in HFB and UFB indicated a probable postcentral shift of the hand motor function in one of the patients. We also compared the task related subband modulations captured by the small and large contacts in our hybrid grid. We did not find any significant difference in terms of band power changes. This study shows initial evidence that event-driven neural oscillatory activity recorded from ECoG can reach up to 800 Hz. The spatial distribution of UFB oscillations was found to be more focalized and closer to the central sulcus compared to LFB and HFB. More studies are needed to characterize further the functional significance of UFB relative to LFB and HFB.

Power Modulations of ECoG Alpha/Beta and Gamma Bands Correlate With Time-Derivative of Force During Hand Grasp

“…Specifically, ERS refers to the power increase, and ERD refers to power decrease during active periods relative to the reference period (Pfurtscheller and Lopes da Silva, 1999). We selected the 8-32 and 60-200 Hz frequency bands based on the visual inspection of the time-frequency maps and they are consistent with our previous studies (Jiang et al, 2017(Jiang et al, , 2018.…”

“…The model of the movement phase is based on a regression analysis that models the initial phase of the bell-shaped EMG power profile as a function of the cubic power of time (Botzer, 2009). This was successfully applied in our previous study to detect the onset of finger flexion and extension (Jiang et al, 2018). As in our recent study (Jiang et al, 2017), we executed a visual inspection using synchronized video, EMG and force sensor to ensure that a detected onset point was a real grasp execution onset rather than an artifact.…”

“…Event-related desynchronization (ERD) in the alpha and beta band (8-32 Hz) and event-related synchronization (ERS) in the gamma band (60-200 Hz) are well-known event-related neural modulations observed in various electrophysiological recordings at both cortical level (Pfurtscheller and Lopes da Silva, 1999;Pfurtscheller et al, 2003;Jiang et al, 2017Jiang et al, , 2018, and subcortical level (Alegre et al, 2005;Brown and Williams, 2005;Kühn et al, 2006) during motor tasks. Specifically, ERS refers to the power increase, and ERD refers to power decrease during active periods relative to the reference period (Pfurtscheller and Lopes da Silva, 1999).…”

“…The statistical significance of ERD/ERS at each channel was tested using a one-tailed Student's t-test with a significance threshold p-value of 0.05 and corrected for multiple comparisons using the False Discovery Rate (FDR) method at the level of 0.05 (Genovese et al, 2002). In order to select the most robust ERD and ERS events, we selected channels that had significant changes of at least 25% from baseline in either grasp onset or offset phase as previously done in (Jiang et al, 2018). For ERD, the alternative hypothesis (H 1 ) was ERD LFB < 0.75 (-1.25 dB), while for ERS, H 1 is ERS HFB > 1.25 (+0.97 dB).…”

“…Brain-machine interfaces (BMI) provide a way to establish voluntary control of prosthetic limbs by decoding signals directly from the sensorimotor area of the brain (Yanagisawa et al, 2011(Yanagisawa et al, , 2012Hochberg et al, 2012;Collinger et al, 2013;Wang et al, 2013;Wodlinger et al, 2015). Event-related desynchronization (ERD) in low frequency band (LFB, and event-related synchronization (ERS) in high frequency band (HFB, are two well-known forms of frequency modulations associated with movements in human (Pfurtscheller and Lopes da Silva, 1999;Miller et al, 2007Miller et al, , 2010Tzagarakis et al, 2010;Su and Ojemann, 2013;Jiang et al, 2017Jiang et al, , 2018 and non-human primates (Sanes and Donoghue, 1993;Ray et al, 2008;Ince et al, 2010;Chen et al, 2014). ERD and ERS modulations have been used successfully in various BMI applications.…”