Electrocardiogram Monitoring Wearable Devices and Artificial-Intelligence-Enabled Diagnostic Capabilities: A Review.

Area of Science:

• Biomedical Engineering
• Artificial Intelligence
• Health Informatics

Background:

• Global population aging and lifestyle changes drive increased incidence of cardiovascular diseases and sleep apnea.
• Wearable devices are advancing for smaller, accurate, and AI-compatible health monitoring.
• Continuous biosignal monitoring enables real-time disease detection and improved patient healthcare management.

Purpose of the Study:

• To review recent advances in using artificial intelligence (AI) with electrocardiogram (ECG) signals from wearable devices and databases.
• To focus on AI-driven detection and prediction of diseases, including heart conditions, sleep apnea, and mental stress.
• To highlight methodological trends in AI for biosignal analysis.

Main Methods:

• Analysis of electrocardiogram (ECG) signals acquired from wearable devices and public databases.
• Application of artificial intelligence (AI) methods, including traditional statistical, machine learning, and advanced deep learning techniques.
• Utilizing deep learning architectures like convolutional neural networks (CNNs) and recurrent neural networks (RNNs) for complex biosignal data.

Main Results:

• Most research focuses on detecting heart diseases, sleep apnea, and emerging areas like mental stress using AI and ECG.
• There is a notable shift towards using advanced deep learning methods over traditional statistical approaches.
• Publicly available databases are predominantly used for developing and validating new AI methods.

Conclusions:

• AI applied to wearable ECG data offers significant potential for early disease detection and prediction.
• Deep learning models are increasingly crucial for analyzing complex biosignal data effectively.
• The trend towards using public datasets facilitates reproducible research and broader AI model development in healthcare.