PCX Dashboard Web App

A lightweight flask web app for comprehensive participant data visualization across a variety of data collection steps, ranging from fMRI analysis to smartphone sensor tracking.

This dashboard links specifically to data within the Holmes Lab Wiki and restricted folders, only accessible to Holmes Lab and PCX study staff.

Note: this web app runs locally on study staff computers, and is not hosted on the web, for privacy of HIPAA data.

Setup

To run this app, first open up your terminal, and clone this directory. git clone https://github.com/kaleyjoss/PCX_dashboard_stable

If you have conda set up, you can create a virtual environment:

conda env create --name dashboard -f environment.yml
conda activate dashboard 

Or, install the following packages with whatever package manager you use:
pandas 2.3.3
numpy 2.3.3
dash 2.14.2 flask 3.0.3 xarray (*This was originally developed with Python 3.13.9)

Or you can install the dependencies from environment.txt file: *Note: only tested on Python 3.13

pip3 install -r requirements.txt

Run app

Then, in the app directory, run this: (--debug is optional, helps with tracebacks and errors)

flask --app app.py run --debug --port 5000

You should have an output which says:

Dash is running on http://127.0.0.1:5000

 * Serving Flask app 'app.py'
 * Debug mode: on

Now navigate to any browser, and go to: "http://127.0.0.1:5000"

You should see the app dashboard there.

Errors and troubleshooting

• If there are any errors which crash the site, you'll see those errors populate in the terminal or the site.

• If there are errors but the site is still running, go to the URL http://127.0.0.1:5000 and check the "Callbacks" popup to see what the errors are.

• If it can't find a specific file/path, or you move/rename a file/folder, change the relevant filepaths in /scripts/paths.py

App Visuals

Homepage

Smartphone Sensor Activity Heatmaps

Daily Survey Interactive Charts

fMRI Visualization and QC Metrics

Interactive Collection Rate Projections