Online browser-based tool for providing an objective hourly hypoglycemia risk score for physical activity decision support leveraging the fixed part of a Mixed Effects Random Forest model (Read more about the model: Modeling risk of hypoglycemia during and following physical activity in people with type 1 diabetes using explainable mixed-effects machine learning ).
The development of the web tool was done with Bokeh, a python library for creating interactive visualizations for web browsers. This tool is publicly available and can be accessed from this link.
Install the following dependencies:
- python=3.8.12
- bokeh=2.4.1
- scikit-learn=1.0.2
Clone the repository
git clone https://github.com/vlt-ro/physical_activity_hypoglycemia_risk_py.git
To run the browser-based tool locally:
bokeh serve GUI_RandomForest
And open the following link in the browser:
localhost:5006/GUI_RandomForest
Distributed under the MIT License. See LICENSE.txt
for more information.