Installation

GlyApp on your Web Browser

On your favorite browser (I recommend Firefox), go to https://jupyter.org/try-jupyter/lab/

Start a new notebook by clicking on Python Pyodide

Copy the following code in the first section. Run the section with Shift+Enter keys or clicking on the button run

import sys, os, re, pyodide, numpy as np, pandas as pd, scipy.ndimage
from datetime import date,time,timedelta,datetime
import matplotlib.pyplot as plt, matplotlib.dates as mdates
url='https://raw.githubusercontent.com/jorishey1234/glyapp/refs/heads/main/gly_toolbox_dev.py'
exec(pyodide.http.open_url(url).read())

After GlyApp is loaded (it can take some time depending on your computer and connection), you should get the answer :

Loading GlyApp dev version
GlyApp successfully loaded !

Hint

You can create as much notebook as you want, for instance one per patient or by study if necessary. Copy the header code all the time to initialize the notebook. When the browser is closed, the notebook is saved automatically in the Browser history. To avoid loosing data, it can be good to externaly save notebooks on a local disk from time to time (Right click on the notebook, then Download).

Start a new section in the notebook, clicking on add .

If running for the first time, prepare the local environment folder structure and files with the command :

>>> init_environment()

Then you should be able to run glycemic statistics

>>> calc_glu('XX')

or get a plot on weekly intervals

>>> plot_patient('XX')

Hint

Any other patient data can now be added in try-jupyter by drag/drop a patient folder containing sensor data in the DATA_DIR, which is by default ./Data/patient/. As for the notebooks, the data is saved automatically when closing the browser, until you clear the history. To avoid loosing data, keep a local version of your DATA_DIR folder, and regularly download the results in the RESULTS_DIR folder (right click on the result file, then Download).

With python (v3.12) installed locally

Make sure you have the following packages installed. For instance with pip

pip install sys, os, re, pyodide, numpy, pandas, scipy, datetime, matplotlib

then, in python run

>>> from glyapp import *
>>> init_environment()
>>> calc_glu('XX')
>>> plot_patient('XX')

GlyApp File Structure

Default Environment Variables

DATA_DIR = './Data/': Path where raw sensor data are kept

RESULT_DIR = './Patients/': Path where Glyapp results are kept

Setting Environment Folders and Files

init_environment(patient='XX')

Prepare the local folder structure for the first time where data has to be stored and results are written.

Create the sensor database if the file do not exist.

Create a test case with synthetic data for patient name XX.

Parameters

patientstr, optional: Codename of the patient to create. The default is ‘XX’.

Returns

None.

make_synthetic_gly_data(patient='XX', ndays=30, dt=15, gly_mean=150, gly_std=150, gly_corr=30, start_date='2025-01-01', seed=0)

Return a CSV file with synthetic glycemic data time series

Parameters

patientstr, optional: Codename of patient to create. The default is ‘XX’.
ndaysint, optional: Duration f the time serie in days. The default is 30.
dtint, optional: Measure intervals in minuts. The default is 15.
gly_meanfloat, optional: Mean glycemia. The default is 150.
gly_stdfloat, optional: Standard deviation of glycemia. The default is 150.
gly_corrfloat, optional: Time correlation of glycemia in min. The default is 30.
start_datestr, optional: Start date of the timeserie. The default is “2025-01-01”.
seedint, optional: Seed for random generator.

Returns

None.