Case studies are a great way to gain experience working with wearables and mHealth data and implementing computational tools from the DBDP. These case studies are part of dbdpED, the educational resource available through the Digital Biomarker Discovery Pipeline.
We have 3 levels of case studies:
- Beginner. You have limited knowledge of Python/R. This tutorial is a step-by-step guide.
- Intermediate. You should know Python/R well and have some experience working with datasets. This tutorial is a step-by-step guide. However, we will give ideas on how you could expand upon the methods in your own work.
- Advanced. You should know Python/R very well and have experience working with datasets in one of the languages. This tutorial is not a step-by-step guide. Rather, it provides the structure and resources for you to explore the data, build your own prediction models, and draw your own conclusions.
- Python(3.0.0+) or R
- Either sufficient space on your personal computing machine to download the data OR ability to work on a cluster. If you have Google Drive, we recommend Google Colab (it's free!).
| Case Study | Level | Language(s) |
|---|---|---|
| Stress | Advanced | Python, R |
| CGM Glucose | Beginner | Python |
| Sleep | Beginner | Python |
| ECG | Beginner | Python |
| more coming soon |
Please refer to individual case studies for questions/issue tracking. If you have an idea or request for a new case study, please open an issue in this repo.