Automatic detection of frontal ECG lead interchange using inter-lead correlation analysis in a 12-lead vital signs monitoring system

Electrode misplacement and interchange remain significant sources of error during electrocardiogram (ECG) acquisition, particularly in portable and connected systems operating within the Internet of Medical Things (IoMT). Such errors can substantially modify ECG morphology, disrupt inter-lead relationships, and mimic pathological cardiac conditions, thereby compromising diagnostic reliability. While commercial devices incorporate proprietary mechanisms for detecting lead reversal, their closed architectures limit transparency and adaptability for research and educational purposes. This study presents an intelligent system for real-time detection of ECG electrode interchange based on inter-lead correlation analysis in standard 12-lead recordings, with limb and augmented leads computed according to conventional clinical definitions. The proposed approach focuses on frontal-plane lead interchange scenarios and exploits short temporal windows to characterize stable correlation patterns associated with specific electrode configurations. The methodology was evaluated using eleven clinically relevant cases involving the frontal leads (I, II, III, aVR, aVL, and aVF). The results demonstrate that inter-lead correlation patterns provide a robust and interpretable means of identifying electrode interchange, allowing technical acquisition errors to be distinguished from physiological signal variations. Beyond supporting acquisition-stage quality control, the proposed framework offers a practical foundation for training applications and provides a scalable basis for future extension toward precordial lead misplacement detection, which is not addressed in the present study.