Automatic ECG Quality Assessment Techniques: A Systematic Review.

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Cardiovascular diseases are the leading global cause of death, with stroke and heart attacks comprising over 80% of related fatalities.
• Accurate diagnosis via electrocardiogram (ECG) analysis is vital for preventing patient mismanagement and saving lives.
• ECG recordings are frequently compromised by noise, which degrades signal quality and complicates diagnosis.

Purpose of the Study:

• To review research on automatic ECG quality assessment techniques published between 2012 and 2022.
• To identify common methodologies, datasets, and performance metrics in automatic ECG quality assessment.
• To discuss limitations in current research and propose recommendations for advancing the field.

Main Methods:

• A systematic review of scientific literature focusing on automatic ECG quality assessment algorithms.
• Analysis of studies utilizing the CinC11 Dataset for training and testing algorithms.
• Evaluation of reported algorithm accuracy rates and assessment methodologies.

Main Results:

• The CinC11 Dataset is the most frequently used resource for algorithm development and validation.
• Most studies evaluated ECG quality on a per-lead basis, with limited real-time testing.
• Reviewed algorithms demonstrated accuracy ranging from 85.75% to 97.15%.

Conclusions:

• Current automatic ECG quality assessment research requires greater methodological clarity for effective clinical implementation.
• Further research should focus on real-time applications and comprehensive subject demographic data.
• Standardized evaluation protocols are needed to improve the reliability and generalizability of ECG quality assessment techniques.