BalancePy

Human balance analysis using Python.

This package implements a series of published modeling approaches, where systems identification techniques are used to identify the neuro-mechanical and sensori-motor mechanisms underlying human balance.

Furthermore, the package integrates with the virtual reality balance analysis software anaropia. Data recorded with anaropia can be directly analyzed using BalancePy.

Installation

pip install balancepy

For development with documentation and testing tools:

pip install balancepy[dev]

For working with Jupyter notebooks:

pip install balancepy[notebooks]

Quick Start

import balancepy

# Access submodules
from balancepy import anaropia, biomechanics, timeseries, frequency
from balancepy import data_class, make_stimulus
from balancepy.model_sim import base_model, parameter

# Load data from anaropia format
com, time = anaropia.getdata_anaropia("path/to/data.csv", output='com')
stim, time = anaropia.getdata_anaropia("path/to/data.csv", output='stimulus')

# Perform frequency domain analysis
freq_data = frequency.analyze(data)

# Access biomechanics calculations
com = biomechanics.get_com(mass_kg=70, height_m=1.75)

Features

• Multi-model framework: Implement and compare different balance control models
• Systems identification: Parameter recovery and model fitting
• Data handling: Standardized time series data structures
• Biomechanics: Common center-of-mass and biomechanics calculations
• Frequency domain analysis: FFT and spectral analysis tools
• Virtual reality integration: Direct support for anaropia data format

Documentation

Full documentation available at: https://posturecontrollab.github.io/balancepy/

References

The package implements modeling approaches from peer-reviewed publications including:

• Peterka, R. J. (2018)
• Assländer, L. (2023)

License

MIT License - See LICENSE file for details

Authors

• Lorenz Assländer (lorenz.asslaender@uni-konstanz.de)
• Matthias Albrecht (matthias.albrecht@uni-konstanz.de)