Oximetry digital biomarkers for the analysis of continuous oximetry (SpO2) time series.
Based on the paper Jeremy Levy, Daniel ́Alvarez, Aviv A Rosenberg, Alexandra Alexandrovich, F ́elix Del Campo, and Joachim ABehar. Digital oximetry biomarkers for assessing respiratory function: standards of measurement, physiologicalinterpretation, and clinical use.NPJ digital medicine, 4(1):1–14, 2021
.. youtube:: 1m0nQ4MIOdE :width: 640 :height: 480
Five types of biomarkers may be evaluated:
-
General statistics: time-based statistics describing the oxygen saturation time series data distribution.
-
Complexity: quantify the presence of long-range correlations in non-stationary time series.
-
Periodicity: quantify consecutive events creating some periodicity in the oxygen saturation time series.
-
Desaturations: time-based measures that are descriptive statistics of the desaturation patterns happening throughout the time series.
-
Hypoxic burden: time-based measures quantifying the overall degree of hypoxemia imposed to the heart and other organs during the recording period.
Available on pip, with the command: pip install pobm
pip project: https://pypi.org/project/pobm/
numpy > 1.18.2
scikit-learn > 0.22.2
scipy > 1.4.1
lempel-ziv-complexity==0.2.2
All the requirements are installed when the toolbox is installed, no need for additional commands.
An example code is available at Notebook.
Available at https://oximetry-toolbox.readthedocs.io/en/latest/