The software is provided for academic research only under the GNU GPL license
Pre-trained model for the diagnosis of obstructive sleep apnea from the nocturnal oximetry time series.
Usage:
OxiNet_API.py [arguments] [options]
Arguments:
<input_signal> Name of the recording to use for inference. The name must be consistent for both sleep stages and oximetry.
Options:
--model_path The path to the pre-trained model for inference.
The default path points to the model that was trained and evaluated as descrived
in the published paper.
Run with default parameters:
$ python3 OxiNet_API.py 354 saved_model
Python 3.7 and above.
numpy==1.21.6
keras==2.11.0
tensorflow==2.11.0
tqdm==4.65.0
joblib==1.2.0
pandas==1.3.5
To install the requirements, use:
$ pip3 install -r requirements.txt
An exe file is also available, to avoid installing the Python environment. To run it, just run the following command line:
$ .\OxiNet_API.exe 354 saved_model