Cardiac arrhythmia detection using deep learning: A review.

Area of Science:

• Cardiology
• Biomedical Engineering
• Artificial Intelligence

Background:

• Electrocardiogram (ECG) analysis is crucial for cardiac arrhythmia detection but faces challenges with massive data volumes and human review limitations.
• Traditional machine learning methods for ECG analysis require manual feature extraction, limiting accuracy due to signal variability and noise.

Purpose of the Study:

• To review recent advancements in deep learning methods for automatic cardiac arrhythmia detection using ECG data.
• To summarize existing literature on deep learning for ECG analysis, covering datasets, applications, input data, model architectures, and performance.

Main Methods:

• Systematic review of recent literature on deep learning approaches for automatic arrhythmia detection.
• Analysis of studies based on utilized datasets, application scope, input data types, model architectures, and performance evaluation metrics.

Main Results:

• Deep learning enables automatic high-level feature extraction and classification, outperforming traditional methods in ECG-based arrhythmia detection.
• The review categorizes and summarizes key aspects of deep learning applications in this field.

Conclusions:

• Deep learning shows significant promise for accurate and automated cardiac arrhythmia detection from ECG.
• Future research should address limitations and explore new opportunities in deep learning for ECG analysis.