This python library contains algorithms for spatiotemporal gait parameter calculation from 3-axis IMU/MARG data (accelerometer, magnetometer, gyroscope).
If you are looking for a more complete project, check out gaitmap: https://github.com/mad-lab-fau/gaitmap
This library is not hosted on PyPi / ReadTheDocs. If you want to use it, install it locally:
pip install -e pigait
To build the sphinx documentation:
cd pigait/docs
sphinx-apidoc.exe -o ./source ../src
make html
Gait event detection methods:
- Wavelet analysis (based on Pham et al 2017)
- Peak detection on gyroscope data (based on Salarian et al 2004 and Mariani et al 2010)
Spatial parameter calculation methods:
- Inverted pendulum model using lumbar data (based on Del Din et al 2016)
- Strapdown inertial navigation / quaternion-based 3D position estimation with zero-velocity updates (ZUPT)
Three combinations have been validated against data simultaneously captured from an optoelectronic motion tracking system.
- Method 1 - Wavelet + inverted pendulum
- Method 2 - Peak detection + inverted pendulum
- Method 3 - Peak detection + strapdown inertial navigation
Results can be found in https://doi.org/10.1016/j.jbiomech.2023.111907
TODO:
- Add examples on open data
- Add tests
- More algorithms etc