Related Experiment Video
Updated: Jun 21, 2026
12:09
Patient-specific Modeling of the Heart: Estimation of Ventricular Fiber Orientations
Published on: January 8, 2013
Modeling day-long ECG signals to predict heart failure risk with explainable AI
Eran Zvuloni1, Ronit Almog2,3, Michael Glikson4
1Faculty of Biomedical Engineering, Technion-IIT, Haifa, Israel. eranzvuloni@gmail.com.
NPJ Digital Medicine
|June 19, 2026
Summary
Artificial intelligence (AI) can predict heart failure (HF) risk using 24-hour electrocardiogram (ECG) data. This non-invasive method identifies high-risk individuals, enabling early intervention to prevent HF hospitalization or death.
Area of Science:
- Cardiology
- Artificial Intelligence
- Medical Informatics
Background:
- Heart failure (HF) significantly impacts quality of life and longevity in older adults.
- Preventing HF is crucial for reducing morbidity and mortality rates.
- Current risk prediction methods may not fully capture dynamic cardiac changes.
Purpose of the Study:
- To investigate the efficacy of artificial intelligence (AI) in predicting heart failure (HF) risk.
- To determine if AI applied to 24-hour single-lead ECG data can forecast HF within five years.
- To compare AI model performance against traditional methods.
Main Methods:
- Utilized the Technion-Leumit Holter ECG (TLHE) dataset (69,663 recordings from 47,729 patients).
- Developed and trained a deep learning model, DeepHHF, on 24-hour ECG recordings.
- Evaluated model performance using area under the receiver operating characteristic curve (AUC) and compared it with shorter ECG segments and clinical scores.
Main Results:
- DeepHHF achieved an AUC of 0.80, outperforming models using 30-second ECG segments and clinical scores.
- High-risk individuals identified by DeepHHF showed a two-fold increased risk of hospitalization or death.
- Explainability analysis indicated DeepHHF focused on arrhythmias and heart abnormalities.
Conclusions:
- Deep learning models can effectively utilize 24-hour continuous ECG data for reliable HF risk prediction.
- AI applied to single-lead Holter ECG is a non-invasive, cost-effective, and accessible tool for HF risk assessment.
- This approach can capture paroxysmal events crucial for accurate risk stratification.
Related Concept Videos
01:29Electrocardiogram
An electrocardiogram (ECG or EKG) is a critical diagnostic tool that records the electrical signals produced by the heart during each heartbeat. This recording is achieved through electrodes placed strategically on the arms, legs, and chest. The electrocardiograph amplifies these signals and produces 12 distinct tracings, offering a comprehensive understanding of the heart's electrical activity.
Three major waveforms are present in a typical ECG recording: the P wave, the QRS complex, and the T...
Three major waveforms are present in a typical ECG recording: the P wave, the QRS complex, and the T...
01:25Correlation between ECG and Cardiac Cycle
The electrical signals recorded on an electrocardiogram (ECG) occur before the mechanical processes of contraction and relaxation during the cardiac cycle.
A cardiac action potential originates in the SA node and spreads throughout the atria and the AV node in approximately 0.03 seconds. This results in the P wave in an ECG and triggers atrial contraction. The action potential is then briefly slowed at the AV node, allowing the atria to contract and fill the ventricles with blood before...
A cardiac action potential originates in the SA node and spreads throughout the atria and the AV node in approximately 0.03 seconds. This results in the P wave in an ECG and triggers atrial contraction. The action potential is then briefly slowed at the AV node, allowing the atria to contract and fill the ventricles with blood before...