Magnetocardiographic Simulation and Analysis of Cardiac QT Interval Disorders

QT interval abnormalities, including prolonged and shortened QT, are linked to life-threatening arrhythmias. While electrocardiogram (ECG) is commonly used for assessment, it has limitations in detecting subtle repolarization changes and spatial dispersion. Magnetocardiography (MCG) offers a potential alternative. We analyzed 735 subjects using MCG, including patients with congenital long QT syndrome (LQTS), short QT syndrome (SQTS), acquired QT prolongation, and healthy controls. QT intervals were measured by automated algorithms and manual annotation, and MCG data were compared to ECG. QTc, defined as the heart rate-corrected QT interval, was calculated for all subjects. A deep learning model was developed for classifying QT disorders. MCG QTc values were consistently longer than ECG values. LQTS patients had significantly prolonged MCG QTc (540±35 ms), while SQTS patients showed shortened MCG QTc (310±20 ms). MCG detected subtle QT prolongation in ECG-normal LQTS patients. The deep learning model achieved 90.8% overall accuracy in classifying QT disorders. MCG effectively measures QT interval abnormalities and identifies subtle repolarization changes missed by ECG, showing promise as a complementary diagnostic tool. MCG, combined with deep learning, provides a noninvasive, accurate approach to diagnosing QT disorders and could enhance personalized treatment strategies.