/ai-clinician-explorer

Primary LanguageSvelte

AI Clinician Explorer

The AI Clinician Explorer (live demo) is a prototype visualization tool and study interface that enables clinicians and health data experts to browse patient trajectories and analyze AI-generated treatment recommendations.

Screenshot of the AI Clinician Explorer showing a patient's demographics, vitals, labs, past treatments, and model predictions.

Sivaraman, V., Bukowski, L., Levin, J., Kahn, J.M., Perer, A. Ignore, trust, or negotiate: Understanding clinician acceptance of AI-based treatment recommendations in health care. ACM CHI 2023.

Setup

pip install -r requirements.txt
cd client && npm install && cd ..

Make sure you have the Firestore key included in the top-level directory.

Start the flask server: python server.py

In a separate terminal, run the frontend: cd client && npm run autobuild

Deployment

In the deployed container, create a secret.txt file using the following command:

python -c 'import secrets; print(secrets.token_hex())' > secret.txt

Create a user profile by running the blueprints/user.py script. Also, make sure you have uploaded the firestore_key.json file.

In the production environment, set the environment variable PRODUCTION_MODE to 1. You can also set a different value for the environment variable FIRESTORE_KEY_PATH to retrieve the key from somewhere other than the default firestore_key.json path.