pyphysio is a library of state of art algorithms for the analysis of physiological signals. It contains the implementations of the most important algorithms for the analysis of physiological data like ECG, BVP, EDA and inertial. The algorithms are implemented on top of a framework that provides caching to optimize feature extraction pipelines.
To allow optimization, two wrapper classes for signals are provided:
- EvenlySignal: wraps a signal sampled with a constant frequency, specifying its values (samples) and its sampling_freq
- UnevenlySignal: wraps a signal where the distance between samples is not constant, specifying the values, the original sampling_freq and the x_values that can be (x_type) 'instants' or 'indices' wrt the original signal.
Every algorithm is available under the main module name e.g.
import pyphysio as ph
ph.IIRFilter(...)
however they are divided into the following groups:
- estimators: from a signal produce a signal of a different type
- filters: from a signal produce a filtered signal of the same type
- indicators: from a signal produce a value
- segmentation: from a signal produce a series of segments
- tools: from a signal produce arbitrary data
Examples on how to use pyphysio can be found at:
https://github.com/MPBA/pyphysio/tree/master/tutorials
If you use pyphysio for research, please cite us:
"Bizzego et al. (2019) 'pyphysio: A physiological signal processing library for data science approaches in physiology', SoftwareX" https://www.sciencedirect.com/science/article/pii/S2352711019301839