The electrocardiographic forward problem: A benchmark study.

Area of Science:

• Biomedical Engineering
• Computational Electrophysiology
• Medical Imaging

Background:

• Electrocardiographic forward problems are key to noninvasive electrocardiographic imaging (ECGI), calculating torso potentials from cardiac sources.
• Despite mature computational methods, experimental validation shows surprisingly high residual errors in forward solutions.
• Potential sources of error in forward problems require further investigation.

Purpose of the Study:

• To investigate the hypothesis that incomplete cardiac source sampling, especially above the atrioventricular (AV) plane, is a major cause of forward solution errors.
• To establish a benchmark for forward computed potentials using comprehensive cardiac source measurements.
• To validate the necessity of complete cardiac source sampling for accurate ECGI.

Main Methods:

• Utilized a modified Langendorff preparation in an electrolytic torso-tank with a novel pericardiac-cage array for thorough cardiac potential sampling.
• Minimized experimental errors by controlling geometry and torso homogeneity.
• Progressively removed signals above the AV plane to quantify the impact of incomplete source sampling on forward-computed torso potentials.

Main Results:

• Complete cardiac source sampling yielded high correlation metrics (≥0.93) between computed and measured torso potentials.
• The mean root-mean-squared error was low (≤0.10 mV) with complete sampling.
• Accuracy of the forward solution dropped significantly when excluding measurements above the AV plane.

Conclusions:

• Complete cardiac source measurements provide a benchmark for future studies.
• This study confirms the critical role of sampling above the AV plane for accurate forward computed torso potentials.
• Improved datasets from complete sampling will enhance inverse techniques and the noninvasive detection of cardiac electrical abnormalities.