Working with Continuous Glucose Monitor (CGM) Data
This tutorial is part of dbdpED, the educational platform for digital biomarker discovery. This tutorial is also available as a Jupyter Notebook. This is a beginner tutorial. If you are more advanced, we recommend our other case studies. Before starting, we recommend that you read this blog on the DBDP and the basics of digital biomarker discovery.
In this case study, we will be using continuous glucose monitor (CGM) data. CGMs are commonly used by people with Type 1 Diabetes.
We will be performing the first few steps of the Digital Biomarker Discovery Pipeline:
1. Pre-processing the data
2. Exploratory data analysis
3. Feature Engineering (using the DBDP cgmquantify module)
Level
Beginner. You have limited knowledge of Python/R. This tutorial is a step-by-step guide.
System Requirements
Python (3.0.0+) — If you don’t yet have Python installed, please follow [this video tutorial](https://www.youtube.com/watch?v=YJC6ldI3hWk) on how to download Python through Anaconda.