Data Driven Computer Simulation to Analyse an ECG Limb Lead System Used in Connected Health Environments

Area of Science:

• Biomedical Engineering
• Cardiology
• Medical Device Development

Background:

• Emerging Connected Health initiatives have led to novel Electrocardiogram (ECG) monitoring devices.
• These devices utilize non-standard limb electrode placements, termed the Central Einthoven (CE) configuration.

Purpose of the Study:

• To compare ECG signals obtained from the CE configuration against the standard Mason-Likar (ML) configuration.
• To evaluate the impact of the CE configuration on ECG signal characteristics and cardiac axis calculation.

Main Methods:

• Computer simulations using body surface potential maps to generate ECGs from both ML and CE configurations (n=176 each).
• Quantitative comparison of ECG signals using root mean square error (RMSE), Pearson correlation (r), and similarity coefficient (SC).
• Analysis of the influence of CE on mean cardiac axis and visual comparison of discrepant cases by an ECG interpreter.

Main Results:

• The CE configuration significantly affects leads aVL, III, and aVF.
• A mean QRS axis shift of 28° towards the right was observed with CE compared to ML.
• CE configuration can lead to emulation of right axis deviation and has an 18% chance of misplacing the axis, with 60% of visually reviewed cases showing clinically significant differences.

Conclusions:

• The CE configuration may compromise diagnostic accuracy for pathologies relying on ECG amplitude criteria.
• Lead II rhythm analysis is unaffected, supporting CE use for rhythm monitoring only.
• Computerized analysis of CE-based ECGs requires careful consideration of these findings to avoid diagnostic errors.