Machine Learning-based Detection of Coronary Artery Disease using Unshielded Magnetocardiography

Magnetic sensing of the heart, or magnetocardiography (MCG), shows promise for detecting ischemia and myocardial infarction. However, clinical adoption has been limited by costly, impractical magnetic shielding and cryogenics. We developed a 26-channel MCG device (SandboxAQ, Palo Alto, CA USA) with optically pumped magnetometers for patients suspected of non-ST elevation acute coronary syndrome (NSTE-ACS) with a bedside form factor. This preliminary report is part of an IRB-approved 150-patient pilot study at Mayo Clinic.

‍