Sugar Sherlock

This is a streamlit-based diabetes tracker and decision support system that predicts glucose levels based on personal insulin and food signals extracted from CGM time series data.

Dashboard

On the dashboard, you can view your CGM data and add the following events:

• Insulin - These are your insulin injections
• Meal - Any meal, food, or snack you eat
• Activities - This can be any physical or psychological activity that effects your glucose levels.
• Fingersticks - Fingerstick blood glucose entries. Useful to keep track accuratacy of the CGM sensor.

Meal Effect Extraction

One key feature of this tool is that you can extract meal effects from your cgm data. This is done by subtracting the insulin effect.

This allows you to build up a library of meal effects that you can learn from and make better dosing/predictions.

Simulator

The simulator is where you can use your extracted meal effects to simulate how you might best dose for a specific meal.

Getting Started

Note: At this time, only Freestyle Libre data is supported.

1. Connect LibreView account

Ensure that your Freestyle Libre is connected to a libreview.com account.

2. Setup secrets.toml

Create the .streamlit/secerts.toml file from the example file:

cp .streamlit/secrets.toml.example .streamlit/secrets.toml

Replace the username and password with your LibreView credentials.

3. Setup virtual environment and install dependencies

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

4. Initialize the database

python app/utils/init_db.py

5. Run the streamlit app

streamlit run app.py