Multi-scale analysis of neural activity in humans: Implications for micro-scale electrocorticography

Kellis, Spencer; Sorensen, L. B.; Darvas, Felix; Sayres, Conor; O’Neill, Kevin; Brown, Richard B.; House, Paul; Ojemann, Jeffrey G.; Greger, Bradley

doi:10.1016/j.clinph.2015.06.002

Cited by 75 publications

(83 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bandwidths of around 60 Hz and 2nd order Butterworth filters offer a good compromise for our application. The frequency spectrum of spikes is usually invariant and does not exceed 3000 Hz (Rey et al, 2015; Kellis et al, 2016). In this work, we defined a frequency range of 100–2000 Hz for the signal analysis which allows to exclude low frequent background signals (below 100 Hz).…”

Section: Spiking Neural Network (Snn) Architecture For Spike Sortingmentioning

confidence: 99%

Spiking Neural Networks Based on OxRAM Synapses for Real-Time Unsupervised Spike Sorting

et al. 2016

View full text Add to dashboard Cite

In this paper, we present an alternative approach to perform spike sorting of complex brain signals based on spiking neural networks (SNN). The proposed architecture is suitable for hardware implementation by using resistive random access memory (RRAM) technology for the implementation of synapses whose low latency (<1μs) enables real-time spike sorting. This offers promising advantages to conventional spike sorting techniques for brain-computer interfaces (BCI) and neural prosthesis applications. Moreover, the ultra-low power consumption of the RRAM synapses of the spiking neural network (nW range) may enable the design of autonomous implantable devices for rehabilitation purposes. We demonstrate an original methodology to use Oxide based RRAM (OxRAM) as easy to program and low energy (<75 pJ) synapses. Synaptic weights are modulated through the application of an online learning strategy inspired by biological Spike Timing Dependent Plasticity. Real spiking data have been recorded both intra- and extracellularly from an in-vitro preparation of the Crayfish sensory-motor system and used for validation of the proposed OxRAM based SNN. This artificial SNN is able to identify, learn, recognize and distinguish between different spike shapes in the input signal with a recognition rate about 90% without any supervision.

show abstract

Section: Spiking Neural Network (Snn) Architecture For Spike Sortingmentioning

confidence: 99%

Spiking Neural Networks Based on OxRAM Synapses for Real-Time Unsupervised Spike Sorting

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Until recently, ECoG electrodes were generally used to identify epileptogenic foci in clinical situations, using large electrodes that typically had a diameter of ~4 mm and an inter-electrode distance of ~10 mm. However, a large number of recent studies have employed methods with higher electrode density, higher channel counts, and smaller electrodes (Rubehn et al, 2009; Ledochowitsch et al, 2011; Matsuo et al, 2011; Toda et al, 2011; Viventi et al, 2011; Escabí et al, 2014; Castagnola et al, 2015; Kellis et al, 2015; Khodagholy et al, 2015; Hotson et al, 2016). While high-density ECoG recording seems to improve BMI performance, for instance, by enhancing naturalistic control of prosthetic arms, few studies have directly demonstrated the efficacy of this technique.…”

Section: Introductionmentioning

confidence: 99%

High Spatiotemporal Resolution ECoG Recording of Somatosensory Evoked Potentials with Flexible Micro-Electrode Arrays

Kaiju

Doi

Yokota

et al. 2017

Front. Neural Circuits

View full text Add to dashboard Cite

Electrocorticogram (ECoG) has great potential as a source signal, especially for clinical BMI. Until recently, ECoG electrodes were commonly used for identifying epileptogenic foci in clinical situations, and such electrodes were low-density and large. Increasing the number and density of recording channels could enable the collection of richer motor/sensory information, and may enhance the precision of decoding and increase opportunities for controlling external devices. Several reports have aimed to increase the number and density of channels. However, few studies have discussed the actual validity of high-density ECoG arrays. In this study, we developed novel high-density flexible ECoG arrays and conducted decoding analyses with monkey somatosensory evoked potentials (SEPs). Using MEMS technology, we made 96-channel Parylene electrode arrays with an inter-electrode distance of 700 μm and recording site area of 350 μm2. The arrays were mainly placed onto the finger representation area in the somatosensory cortex of the macaque, and partially inserted into the central sulcus. With electrical finger stimulation, we successfully recorded and visualized finger SEPs with a high spatiotemporal resolution. We conducted offline analyses in which the stimulated fingers and intensity were predicted from recorded SEPs using a support vector machine. We obtained the following results: (1) Very high accuracy (~98%) was achieved with just a short segment of data (~15 ms from stimulus onset). (2) High accuracy (~96%) was achieved even when only a single channel was used. This result indicated placement optimality for decoding. (3) Higher channel counts generally improved prediction accuracy, but the efficacy was small for predictions with feature vectors that included time-series information. These results suggest that ECoG signals with high spatiotemporal resolution could enable greater decoding precision or external device control.

show abstract

“…In most cases, these high density arrays have covered a very small area of cortex, and therefore large scale comparisons of shared signal in space have not been evaluated. Recently, correlations in the raw signal have been explored with respect to distance using a clinical ECoG array and a microwire ECoG array(Kellis et al, 2015). However, the measured signals from the microwire array are inherently different from those of the clinical array due to filtering effects of electrode size and shape(Kent & Grill, 2014; Lempka et al, 2011).…”

Section: Introductionmentioning

confidence: 99%

Spatial resolution dependence on spectral frequency in human speech cortex electrocorticography

et al. 2016

View full text Add to dashboard Cite

Objective Electrocorticography (ECoG) has become an important tool in human neuroscience and has tremendous potential for emerging applications in neural interface technology. Electrode array design parameters are outstanding issues for both research and clinical applications, and these parameters depend critically on the nature of the neural signals to be recorded. Here, we investigate the functional spatial resolution of neural signals recorded at the human cortical surface. We empirically derive spatial spread functions to quantify the shared neural activity for each frequency band of the electrocorticogram. Approach Five subjects with high-density (4mm center-to-center spacing) ECoG grid implants participated in speech perception and production tasks while neural activity was recorded from the speech cortex, including superior temporal gyrus, precentral gyrus, and postcentral gyrus. The cortical surface field potential was decomposed into traditional EEG frequency bands. Signal similarity between electrode pairs for each frequency band was quantified using a Pearson correlation coefficient. Main results The correlation of neural activity between electrode pairs was inversely related to the distance between the electrodes; this relationship was used to quantify spatial falloff functions for cortical subdomains. As expected, lower frequencies remained correlated over larger distances than higher frequencies. However, both the envelope and phase of gamma and high gamma frequencies (30-150Hz) are largely uncorrelated (<90%) at 4mm, the smallest spacing of the high-density arrays. Thus, ECoG arrays smaller than 4mm have significant promise for increasing signal resolution at high frequencies, whereas less additional gain is achieved for lower frequencies. Significance Our findings quantitatively demonstrate the dependence of ECoG spatial resolution on the neural frequency of interest. We demonstrate that this relationship is consistent across patients and across cortical areas during activity.

show abstract

Multi-scale analysis of neural activity in humans: Implications for micro-scale electrocorticography

Abstract: Recorded with appropriately scaled electrodes and grids, field potentials expose a more detailed representation of cortical network activity, enabling advanced analyses of cortical pathology and demanding applications such as brain-computer interfaces.

Cited by 75 publications

References 82 publications

Spiking Neural Networks Based on OxRAM Synapses for Real-Time Unsupervised Spike Sorting

Spiking Neural Networks Based on OxRAM Synapses for Real-Time Unsupervised Spike Sorting

High Spatiotemporal Resolution ECoG Recording of Somatosensory Evoked Potentials with Flexible Micro-Electrode Arrays

Spatial resolution dependence on spectral frequency in human speech cortex electrocorticography

Contact Info

Product

Resources

About