¶ Supported Operators and Algorithms in Instinct
Nexstem's advanced EEG headset Instinct is equipped with a comprehensive suite of operators and algorithms designed to enhance the analysis and processing of brainwave data. This array of tools allows researchers and clinicians to apply complex filters, perform data transformation, and utilize statistical analysis to extract meaningful insights from EEG signals. Here, we explore the supported filters, data-processing operators, and algorithms that make Nexstem a versatile and powerful tool in neurotechnology.
¶ Filter Operators
¶ IIR (Infinite Impulse Response) Filter
IIR filters are used for their efficiency in real-time applications, providing a recursive approach to filter EEG signals. They are particularly effective in reducing high-frequency noise and are configurable to achieve desired frequency response with precision.
¶ FIR (Finite Impulse Response) Filter
FIR filters are non-recursive and inherently stable, making them suitable for precisely controlling the phase characteristics of EEG signals. They are widely used for their linear phase response, crucial for maintaining the temporal relationships within EEG data.
¶ AWT (Adaptive Wavelet Thresholding) Filter
The AWT filter uses wavelets to adaptively denoise EEG signals, effectively preserving signal characteristics while reducing noise. This method is beneficial for handling non-stationary signals like EEG, where signal frequency components vary over time.
¶ ASR (Artifact Subspace Reconstruction) Filter
ASR is a sophisticated technique for removing artifacts from EEG data. By reconstructing the signal subspace without the components that are considered artifacts, ASR ensures the preservation of true brainwave signals even in the presence of significant external noise.
¶ Data-Processing/Data-Transformation Operators
¶ EMD (Empirical Mode Decomposition)
EMD is a powerful technique for decomposing a signal into its inherent modes of oscillation, facilitating analysis of complex data collected from non-linear and non-stationary processes.
¶ FastICA (Independent Components Analysis)
FastICA is an algorithm that performs independent component analysis, which is crucial for separating mixed signals into their constituent components. It’s particularly useful in EEG analysis for artifact removal and source localization.
¶ WT (Wavelet Transform)
Wavelet Transform provides a time-frequency representation of the EEG signal, enabling the analysis of different frequency components with respect to time—essential for studying brain dynamics.
¶ Statistical Operators
These include calculations like mean, median, skewness, and more. Statistical operators are vital for summarizing EEG data, facilitating the understanding of its distribution and central tendencies.
¶ Resampling Operator
Resampling is used to change the sampling rate of EEG data, which can be crucial for aligning datasets or preparing data for further analysis.
¶ DFT (Discrete Fourier Transform)
DFT transforms time-domain signals into their frequency components, essential for frequency analysis in EEG studies.
¶ PSD (Power Spectral Density) and Welch PSD
These methods estimate the power of a signal at different frequencies, helping in the analysis of spectral density over time.
¶ Bandpower Operator
This operator calculates the power of the signal in a specific frequency band, commonly used in EEG analysis to quantify the energy of brain waves within specific frequency ranges.
¶ Advanced Analytical Algorithms
¶ CCA (Canonical Correlation Analysis)
CCA is used to identify the relationships between two sets of variables. It is effective in EEG to study the correlation between different regions of the brain.
¶ LDA (Fisher’s Linear Discriminant Analysis)
LDA is used for pattern classification and has been effectively applied in BCI to differentiate between different mental states based on EEG data.