Non-invasive mapping of action potential waveform reconstructed from unshielded Magnetocardiography. Update and limitations

Objective: To explore the feasibility and limitations of reconstructing ventricular action potential waveforms from non-invasive, unshielded magnetocardiographic mapping (uMCG), highlighting significant differences between healthy individuals and patients, even at the current precision level. Methods: uMCG was performed using a 36-channel DC-SQUID system. Mathematical modeling and the calculation method proposed by Kandori et al. were employed to reconstruct the ventricular action potential (RAP) from cardiac magnetic field data of 10 healthy volunteers and 10 patients with various cardiomyopathies. Four patients had simultaneous uMCG and monophasic action potential (MAP) recordings using an amagnetic catheter. Results: Magnetic reconstruction of ventricular action potentials was feasible in all subjects, and averages of 90-second uMCG recordings were sufficient to obtain RAPs comparable to those from 300-second recordings. The RAP waveform aligned closely with that of the MAP at approximately 50% of mapping sites. Significant prolongation of the RAP’s phase-0 (p < 0.01) and a trend toward longer RAP duration (p = 0.06) were observed in patients compared with healthy individuals, underscoring the sensitivity of this technique in detecting electrophysiological abnormalities. Conclusions: Although incomplete RAP repolarization at certain sites points to limitations in spatial resolution, the reproducibility of results suggests that short-duration MCG recordings may be sufficient for reliable action potential reconstruction in unshielded clinical settings. This strengthens the case for MCG as a promising non-invasive medical device for multipurpose cardiac electrophysiological studies. Future advancements in sensor density, refined mathematical models, and multimodal imaging could enable quasi-cellular precision, further bridging the gap between experimental and clinical electrophysiology.