Related Experiment Video
Updated: Nov 12, 2025
In Silico Clinical Trials for Cardiovascular Disease
Published on: May 27, 2022
Machine Learning for Real-Time Heart Disease Prediction
Insights
This study introduces a new machine learning method for real-time detection of heart anomalies using electrocardiograms (ECGs). The model accurately identifies 7 types of heart rhythm signals, improving diagnostic capabilities.
Area of Science:
- Cardiology
- Machine Learning
- Signal Processing
Background:
- Heart anomalies are a leading cause of global mortality, often presenting without symptoms.
- Accurate identification of cardiac anomalies requires specialized personnel and can be challenging.
- Digital Electrocardiograms (ECGs) offer a rich data source for developing automated diagnostic tools.
Purpose of the Study:
- To develop and validate a novel machine learning methodology for real-time ECG analysis.
- To accurately classify seven distinct types of cardiac rhythm signals.
- To achieve high diagnostic performance across diverse datasets and recording standards.
Main Methods:
- Feature extraction from ECG signals.
- Application of the XGBoost machine learning algorithm for model training.
- Validation using a dataset of nearly 40,000 expert-labeled ECGs from multiple institutions and countries.
Main Results:
- The developed models achieved high out-of-sample F1 Scores ranging from 0.93 to 0.99.
- Real-time prediction capability with analysis completed in under 30 milliseconds.
- Successful detection of Normal, Atrial Fibrillation (AF), Tachycardia, Bradycardia, Arrhythmia, Other, and Noisy ECG signals.
Conclusions:
- The proposed XGBoost-based methodology demonstrates exceptional performance in detecting various ECG anomalies.
- This approach offers a robust and efficient solution for real-time cardiac anomaly detection.
- The study represents a significant advancement, achieving high performance across diverse clinical settings and recording standards.
Abstract:
Heart-related anomalies are among the most common causes of death worldwide. Patients are often asymptomatic until a fatal event happens, and even when they are under observation, trained personnel is needed in order to identify a heart anomaly. In the last decades, there has been increasing evidence of how Machine Learning can be leveraged to detect such anomalies, thanks to the availability of Electrocardiograms (ECG) in digital format. New developments in technology have allowed to exploit such data to build models able to analyze the patterns in the occurrence of heart beats, and spot anomalies from them. In this work, we propose a novel methodology to extract ECG-related features and predict the type of ECG recorded in real time (less than 30 milliseconds). Our models leverage a collection of almost 40 thousand ECGs labeled by expert cardiologists across different hospitals and countries, and are able to detect 7 types of signals: Normal, AF, Tachycardia, Bradycardia, Arrhythmia, Other or Noisy. We exploit the XGBoost algorithm, a leading machine learning method, to train models achieving out of sample F1 Scores in the range 0.93 - 0.99. To our knowledge, this is the first work reporting high performance across hospitals, countries and recording standards.
More Related Videos
Related Concept Videos
01:30Pulse rhythm
Conversely, an irregular pulse pattern is termed dysrhythmia, stemming from disruptions in cardiac...
01:30Factors Influencing Heart Rate
Let us explore the significant factors affecting heart rate, including age, body temperature, posture, acute pain, chemical influences,...
01:30Heart Failure IV: Classification and Diagnostic Evaluation