neurokit2

NeuroKit2 is a robust, open-source Python framework designed for the comprehensive analysis of physiological data, often referred to as biosignals. It serves as an essential tool for researchers in fields such as psychophysiology, neuroscience, clinical diagnostics, and human-computer interaction (HCI). By providing a standardized, modular, and highly extensible codebase, NeuroKit2 enables users to transform raw sensor data into actionable clinical or scientific insights across multiple domains of autonomic and central nervous system monitoring.

• Cardiac Signal Processing: Offers full-scale pipelines for ECG and PPG, including signal cleaning, R-peak detection, morphology delineation, and quality assessment.

• Heart Rate Variability (HRV) Analysis: Computes extensive metrics across time, frequency, and nonlinear domains, such as SDNN, RMSSD, Poincaré plots, and entropy measures.

• Brain Activity Analysis: Supports EEG frequency band power calculation (Alpha, Beta, Delta, Theta, Gamma), microstate segmentation, global field power analysis, and MNE-Python integration.

• Autonomic Nervous System Assessment: Decomposes Electrodermal Activity (EDA/GSR) into tonic and phasic components to identify skin conductance responses and sympathetic nervous system indices.

• Respiratory & Muscular Analysis: Processes RSP signals to extract breathing rate and variability (RRV), and handles EMG data for muscle activation detection and amplitude quantification.

• Eye Movement Processing: Provides EOG analysis capabilities, including blink detection and feature extraction for eye tracking studies.

• Multi-modal Integration: Facilitates the synchronized analysis of diverse physiological inputs, allowing researchers to study the interaction between cardiovascular, neural, and respiratory systems.

• Advanced Signal Utilities: Includes broad-purpose tools for filtering, resampling, power spectral density estimation, wavelet decomposition, and autocorrelation.

• Use this skill to automate end-to-end processing workflows for large psychophysiological datasets, reducing manual intervention and ensuring procedural consistency.

• Input data should typically be in array or dataframe format with defined sampling rates; outputs are generally returned as structured dataframes and informational metadata dictionaries.

• Always verify signal quality and filter noise prior to feature extraction to maintain the statistical validity of heart rate variability or neural power metrics.

• Leverage the built-in integration with established libraries such as Pandas, NumPy, and Scipy for seamless data manipulation in research pipelines.

• Ideal for academic researchers conducting event-related potential (ERP) studies, clinical researchers monitoring patients in ICU, or engineers developing biosensing wearables.