Broken Heart - Whoop Health Data Comparison

A health data analysis application that compares Whoop wearable metrics with clinical lab test results during critical cardiac events.

Overview

This application integrates Whoop health data (recovery, sleep, HRV, heart rate) with lab test results to analyze biomarker correlations during a critical health period that included:

• January 11, 2026: Cardiac arrest during tennis. Immediate bystander CPR, Vfib for EMS, ROSC with defib x1. Remarkably neurologically intact & conversational immediately after event; Was not intubated.
• January 19, 2026: Triple bypass surgery

The analysis covers January 1 - February 10, 2026.

🚀 Quick Start - Streamlit Cloud (Recommended)

Deploy for free in 5 minutes!

1. Fork or clone this repository
2. Go to share.streamlit.io
3. Deploy from your GitHub repo (select dashboard.py)
4. Configure secrets in Streamlit Cloud dashboard
5. Update Whoop OAuth redirect URI Features 📖 Full deployment guide: See DEPLOYMENT.md

Local Development

• Whoop OAuth 2.0 Integration: Secure authentication with Whoop API
• Automated Data Fetching: Retrieves recovery, sleep, cycle, and workout data
• Lab Test PDF Parsing: Extracts biomarker values from lab result PDFs
• Interactive Dashboard: Streamlit-based visualization with event markers
• Correlation Analysis: Compare Whoop metrics with clinical biomarkers
• Timeline Visualization: Track health metrics around critical events

Setup (Local Development Only)

For cloud deployment, see DEPLOYMENT.md

1. Install Dependencies

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

For Streamlit Cloud: Configure secrets in the Streamlit Cloud dashboard (see DEPLOYMENT.md)

For local development

Copy the example environment file and add your Whoop credentials:

For local development:

# Option A: Use Streamlit secrets
cp .streamlit/secrets.toml.example .streamlit/secrets.toml
# Edit .streamlit/secrets.toml with your credentials

# Option B: Use .env file
cp .env.example .env
# Edit .env with your credentials

Configuration needed:

• WHOOP_CLIENT_ID: Your Whoop app client ID
• WHOOP_CLIENT_SECRET: Your Whoop app client secret
• WHOOP_REDIRECT_URI: For local use http://localhost:8501, for cloud use your Streamlit URL

3. Run the Dashboard

streamlit run dashboard.py

Visit http://localhost:8501

Authentication (Local & Cloud)

1. Navigate to "Data Management" page
2. Click "🔐 Login with Whoop"
3. Authorize the application
4. You'll be redirected back authenticated

Cloud Deployment

Deploy to Streamlit Community Cloud (FREE):

1. Push code to GitHub
2. Visit share.streamlit.io
3. Create new app from your repo
4. Configure secrets in Streamlit dashboard
5. Update Whoop OAuth redirect URI

Complete deployment guide: DEPLOYMENT.md

Why Streamlit Cloud?

✅ Free hosting for unlimited public apps
✅ Automatic updates when you push to GitHub
✅ Built-in secrets management
✅ No server management required
✅ HTTPS included for OAuth security

Usage (Local or Cloud)

Fetch Data

1. Authenticate with Whoop (via Data Management page)
2. Click "Fetch Whoop Data" to retrieve Jan 1-Feb 10 metrics
3. Click "Parse Lab PDFs" to extract lab test results

View Dashboards

• Overview: Critical events timeline and data status
• Detailed Metrics: Individual metric charts with event markers
• Lab Correlations: Compare Whoop vs clinical biomarkers

Data Persistence

Cloud Deployment: Streamlit Cloud uses ephemeral storage. Data is lost on app restart.

• Solution: Re-fetch data when needed (respects Whoop rate limits)
• Alternative: Add external storage (S3, GCS) for production use

Local Development: Data persists in data/ directory

1. Go to "Data Management" page
2. Click "Fetch Whoop Data" to retrieve metrics from Jan 1 - Feb 10
3. Click "Parse Lab PDFs" to extract data from PDF files in health-data/

Project Structure

broken-heart/
├── config.py              # Configuration and settings (supports Streamlit secrets)
├── whoop_auth.py          # OAuth 2.0 authentication handler
├── whoop_client.py        # Whoop API client
├── streamlit_oauth.py     # Streamlit-native OAuth flow (no Flask needed)
├── lab_parser.py          # PDF parser for lab results
├── data_processor.py      # Data processing and storage
├── dashboard.py           # Streamlit visualization dashboard
├── requirements.txt       # Python dependencies
├── DEPLOYMENT.md          # Streamlit Cloud deployment guide
├── .env.example           # Environment variables template (local dev)
├── .streamlit/
│   ├── config.toml        # Streamlit configuration
│   └── secrets.toml.example  # Secrets template
├── health-data/           # Lab test PDF files
├── data/                  # Processed data output (ephemeral on cloud)
└── tokens/                # OAuth tokens (not in git)

Data Files

After fetching and processing, the following files are created in data/:

• daily_metrics.csv: Combined daily Whoop metrics
• lab_data.csv: Parsed lab test results
• whoop_raw_data.json: Raw API responses
• lab_raw_data.json: Raw parsed lab data

Whoop Metrics Tracked

• Recovery: Recovery score, HRV (RMSSD), resting heart rate, SpO2, skin temperature
• Sleep: Performance, efficiency, sleep stages (light, deep, REM), respiratory rate
• Activity: Day strain, average/max heart rate, energy expenditure
• Workouts: Sport type, workout strain, heart rate metrics

Lab Tests Supported

• Basic Metabolic Panel (glucose, sodium, potassium, creatinine, etc.)
• Complete Blood Count (WBC, RBC, hemoglobin, hematocrit, platelets)
• Glucose monitoring

Correlation Analysis

The dashboard highlights potential correlations:

• HRV vs Metabolic Markers: Lower HRV may indicate metabolic stress

• Recovery Score vs CBC: Drops during illness/inflammation

• SpO2 vs Hemoglobin: Oxygen transport capacity

• Never commit secrets: .env and .streamlit/secrets.toml are in .gitignore

• Use Streamlit Secrets UI: For cloud deployment, only configure secrets via Streamlit dashboard

• OAuth tokens: Stored locally, never committed to git

• HTTPS required: Whoop OAuth requires HTTPS (Streamlit Cloud provides this automatically)

• Never commit .env file or tokens/ directory

• OAuth tokens are encrypted and stored locally

• Use strong FLASK_SECRET_KEY in production

Troubleshooting

PDF Parsing Issues

If automated PDF parsing fails, you can manually enter data:

from lab_parser import create_manual_lab_data_template
template = create_manual_lab_data_template()
# Edit template and save to data/lab_data.json

Authentication Issues

• Ensure redirect URI matches exactly in Whoop developer dashboard
• Check that all required OAuth scopes are granted
• Token refresh happens automatically but can be forced by re-authenticating

Rate Limits

Whoop API has rate limits (100 req/min). The client handles pagination automatically and tracks rate limit headers.

License

Personal use project for health data analysis.