New AI Model Predicts Atrial Fibrillation 30 Minutes Before Occurrence

Researchers at the University of Luxembourg develop a deep-learning AI that forecasts the onset of atrial fibrillation, enhancing early intervention possibilities.

• The AI model, named WARN, uses heart rate data to predict atrial fibrillation with around 80% accuracy.
• WARN can be integrated into wearable devices like smartwatches, allowing continuous monitoring and early warnings.
• The model was trained on 24-hour ECG recordings from 350 patients at Tongji Hospital, China.
• Future developments include personalizing the model to improve prediction times based on individual heart dynamics.
• The technology aims to reduce emergency interventions and improve patient outcomes by enabling early preventive measures.

3 Articles

ScienceDaily NDTV New Atlas