acti-data-analyzer 1.0.3

A library for processing and analyzing actigraphic data.

ADA – Actigraphic Data Analyzer

ADA stands for Actigraphic Data Analyzer, coincidentally being also the name of the first programmer in the world. It is an open source Python module for processing, analyzing and visualizing actigraphic data, with the main focus on bedtime sleep/wake classification and circadian rhythms analysis. ADA is usable in form of a python package, and in form of standalone application with graphical interface (GUI).

Installation

Python 3.11 or higher is required. To install the Python package use pip:

pip install acti-data-analyzer

GUI version of ADA can installed and run on any system as a Python script, best within a venv. To do use following commands:

pip install 'acti-data-analyzer[gui]'
python -m ada_gui

Main features

• Reading and writing data from:
  • GENEActiv .bin and .csv files
  • Actigraph Corp .gt3x and .csv files
  • The MESA dataset.
• Converting data to native .ada format (much more efficient in terms of disk space and reading speed).
• Multiple algorithms allowing collapsing into epochs:
  • MIMS
  • ActivityIndex
  • ENMO
  • Resampling
• Sleep/wake scoring using multiple algorithms:
  • Cole-Kripke
  • Webster
  • Scripps Clinic
  • UCSD
  • Sazonov
  • Unified Filter, with possibility of custom filter construction.
• Estimating multiple sleep metrics:
  • Sleep efficiency
  • Sleep fragmentation index
  • Sleep onset latency
  • Wake after sleep onset
  • And more.
• Assessing circadian rhythms:
  • Single and multi-component cosinor in linear and nonlinear variants
  • Sigmoidally-transformed cosinor
  • Spectrum estimator using autoregressive model
  • Detrended Fluctuation Analysis
  • Nonparametric measures: Interdaily Stability, Intradaily Variability, M10, L5.
• Easy to read and analyze summaries of sleep/wake and circadian assessments for multiple subjects (in a form of comma-separated CSV).
• Automatic and customizable synchronization of actigraphic data with PSG staging.
• Plotting of all analysis steps.

Documentation

Full documentation together with tutorial providing basic insight into functionalities of ADA as a Python package can be found under the link.

Example data

Sample dataset, consisting of 87 weekly recordings of healthy subjects, can be found under the link. Do not hesitate to test the package capabilities using this data!

Acknowledgments

pygt3x source code included in ADA is developed under the GPL-3.0 license by the Actigraph team. The original repository can be found under the link.

Funding

Study financed from the state budget within the program of the Polish Minister of Education and Science under the name ”Perły Nauki”, project number PN/01/0111/2022, funding value 239 998.00 zł, total value 239 998.00 zł