Electrocardiology 2014 - Proceedings of the 41 st International Congress on Electrocardiology 65 Modified Lewis ECG Lead System for Ambulatory Monitoring of Atrial Arrhythmias 1 A. Petrơnas, 1 V. Marozas, 2 L. Srnmo, 3 G. Jaru!eviþius, 1 D. Gogolinskaitơ 1 Biomedical Engineering Institute, Kaunas University of Technology, Kaunas, Lithuania 2 Department of Biomedical Engineering, Lund University, Lund, Sweden 3 Institute of Cardiology, Lithuanian University of Health Sciences, Kaunas, Lithuania Email: andrius.petrenas@ktu.lt Abstract. The analysis of atrial activity (AA) during atrial arrhythmias can be problematic when a reduced lead system is used due to low amplitude and noise. Although leads for AA enhancement were proposed many years ago by Sir Thomas Lewis, two electrodes need to be placed directly on the chest, and therefore arm movement artefacts are likely to occur. In this study, we propose a modified Lewis lead system better suited for ambulatory applications where the electrodes are placed in areas with less muscle. The proposed modification was compared to the Lewis leads as well as to the ES lead of the EASI system. Forty-one healthy volunteers and 8 patients with atrial fibrillation participated in the study. The results show that the proposed lead exhibits the best atrial-to-electromyographic activity ratio, with twice as large AA amplitude as the original leads. Furthermore, the atrial-to-ventricular activity ratio is 50% better than that of the ES lead. The results suggest that the proposed modification of the Lewis lead system has a potential to improve ambulatory monitoring of atrial arrhythmias. Keywords: Atrial activity enhancement; atrial fibrillation; Lewis ECG lead system 1. Introduction While it is well-known that atrial fibrillation (AF) is a progressive disease, with brief episodes evolving into longer and eventually persistent, a recent debate has arisen whether brief episodes of AF are related to cryptogenic ischemic stroke [1], [2]. The hypothesis that brief AF episodes (< 30 s) can contribute to thrombus formation has yet to be proven, and thus automatic detection of very short episodes could accelerate more accurate diagnosis. The problem of false alarms due to electromyographic (EMG) noise, motion artefacts, and ectopic beats of commercial equipment for AF detection [3] forces cardiologists to review software- defined arrhythmia episodes manually, especially if AF events are brief [1], [2]. Since manual revision is exceedingly time-consuming and sometimes unreliable [4], increased accuracy of AF detection devices is required to ensure that brief AF events are detected in long-term ECG recordings. Since most algorithms for AF detection are based solely on the analysis of RR interval irregularity, the many false positives still represent an unsolved problem. While attempts have been made to reduce the number of false positives by involving information on atrial activity (AA) in the AF detection process, the performance of such algorithms has turned out to not be better than those based on RR interval information [3]. The main reason for this outcome is that the conventional 12-lead ECG system, as well as reduced-lead modifications, are focused on ventricular activity (VA), and thus the electrode placement is not optimal for analysing AA. Due to the fact that AA amplitude is small compared to VA, an ECG lead with increased AA amplitude is beneficial to better discriminate between various arrhythmias of atrial origin (atrial tachycardia / flutter / fibrillation), as well other arrhythmias such as wide QRS complex tachycardia [5].