Heart Digital Twin

Overview

The Heart Digital Twin is a real-time cardiovascular monitoring system that uses ECG data from the MIT-BIH Arrhythmia Database. It shows live ECG plots with anomaly detection, alongside an interactive 3D heart model that beats in time with the cardiac rhythm.

Example Structure

The Heart Digital Twin has these components:

• ECG Data Processing: Real-time processing and visualisation of electrocardiogram signals from the MIT-BIH database
• 3D Heart visualisation: Interactive 3D heart model with synchronised pumping animation
• Anomaly Detection: Real-time identification and highlighting of cardiac anomalies
• Web Interface: Flask-based web application for real-time monitoring and control

MIT-BIH Arrhythmia Database

This digital twin uses the MIT-BIH Arrhythmia Database, which contains 48 half-hour excerpts of two-channel ambulatory ECG recordings. The database includes:

• Records from 47 subjects studied by the BIH Arrhythmia Laboratory
• 360 Hz sampling frequency with 11-bit resolution
• Expert annotations for arrhythmia detection and classification
• Multiple ECG channels (MLII, V1, V2, V4, V5) depending on the record

Records 100-109 are available for simulation. Each contains different types of cardiac rhythms and anomalies.

Digital Twin Configuration

This example uses the following assets:

Asset Type	Names of Assets	Visibility	Reuse in Other Examples
Data	MIT-BIH Arrhythmia Database Records (100-109)	Public	Yes
Models	3D Heart Model (Beating heart.glb)	Public	Yes
Functions	ECG Processing & Anomaly Detection	Public	Yes
Tools	Flask Web Server & UI Components	Public	Yes

You can customise the heart digital twin with these parameters:

• Record Selection: Choose from records 100-109
• Channel Selection: Switch between available ECG channels
• Window Size: Adjust the time window for real-time display
• Update Frequency: Control the refresh rate of the simulation

3D Heart Animation

The digital twin includes a synchronised 3D heart pumping animation:

• Real-time Synchronisation: Heart model animation matches the ECG rhythm
• Interactive Controls:
• Camera controls for 360° viewing
• Auto-rotate functionality
• Reset view option
• Animation Features:
• Realistic beating heart model
• Smooth cardiac cycle animation
• Pause/resume functionality synchronised with ECG simulation
• Model Format: GLB (GL Transmission Format Binary) 3D model
• Rendering: WebGL-based rendering using model-viewer component

The 3D heart moves in step with the electrical activity shown in the ECG, which helps to relate the electrical signals to the mechanical function of the heart.

Lifecycle Phases

Lifecycle Phase	Completed Tasks
Create	Sets up Python virtual environment and installs required dependencies (NumPy, Pandas, Matplotlib, WFDB, Flask)
Execute	Starts the Flask web server and launches the heart monitoring interface accessible via web browser
Clean	Terminates the running application and removes the virtual environment

Run the example

To run the example, go to the following directory:

cd /workspace/examples/digital_twins/heart_dt

If required, make the lifecycle scripts executable:

chmod +x lifecycle/create
chmod +x lifecycle/execute
chmod +x lifecycle/clean

Now, run the following scripts:

Create

Sets up the Python environment and installs the required dependencies, including WFDB for MIT-BIH data access, Flask for the web interface, and scientific computing libraries.

lifecycle/create

Execute

Starts the Heart Digital Twin web application in the background. Open it in a web browser for real-time ECG monitoring and 3D heart visualisation.

lifecycle/execute

The application is available at http://localhost:5001. On start-up it:

• Loads the required MIT-BIH records
• Initialises the ECG data processing
• Starts real-time visualisation
• Enables 3D heart model synchronisation

Examine the results

Use the web interface to:

• Monitor Real-time ECG: View live electrocardiogram signals
• Control Simulation: Select different records, channels, and update frequencies
• Interact with 3D Heart: Rotate, zoom, and control the heart animation
• Detect Anomalies: Observe highlighted cardiac irregularities
• Pause/Resume: Control the simulation flow

The application writes its logs to the logs/heart_dt.log file.

Clean

Stops the running application and cleans up the environment.

lifecycle/clean

This script:

• Stops the Flask web server
• Removes the Python virtual environment

References

More information about the MIT-BIH Arrhythmia Database and cardiac monitoring:

Moody GB, Mark RG. The impact of the MIT-BIH Arrhythmia Database.
IEEE Eng in Med and Biol 20(3):45-50 (May-June 2001).

Goldberger, A., et al. PhysioBank, PhysioToolkit, and PhysioNet:
Components of a new research resource for complex physiologic signals.
Circulation 101(23):e215-e220, 2000.

The MIT-BIH Arrhythmia Database is available at: https://physionet.org/content/mitdb/