Meta-Analysis of Blood Glucose Prediction Models and Methods for Meal and Physical Activity Detection in the Context of Diabetes Mellitus

This GitHub repository contains the code and data related to the academic article Meta-Analysis of Technologies for Diabetes Treatment: Glycemic Control, Prediction, Meal and Physical Activity Detection, published in the Progress in Biomedical Engineering journal and available via the IOPscience link below. Please, cite the linked paper if you want to reference this repository in your works.

How to Use

To use this code, you first need to install Python on your computer and optionally some IDE in which you can comfortably interact with the scripts. Specifically, I have used Python version 3.11.4 and the following libraries:

LIBRARY	VERSION
matplotlib	3.7.2
numpy	1.25.2
pandas	2.1.0
scikit-learn	1.3.0
scipy	1.11.2

With everything prepared, the code should be ready to use.

Description of Files

The repsitory contains the following files:

1. plots.py – implements the meta-analysis, i.e., creating the plots and printing reported values;
2. algorithms.csv – data related to the Time-Shift, Pattern Prediction, and BGLP 2025 winnter models;
3. meal.csv – data related to the meal detection models;
4. physicalactivity.csv – data related to the physical activity detection models;
5. prediction.csv – data related to the blood glucose prediction models;