Assessing the accuracy of limited lead recordings for the detection of Atrial Fibrillation

Atrial Fibrillation (AF) is a common cardiac arrhythmia which, if left untreated, can lead to ischaemic stroke. A number of low cost hand held devices which use a single lead I ECG to facilitate detection of a cardiac arrhythmia have recently appeared on the market. This study aimed to assess the accuracy of using limited lead ECG recordings for the detection of AF using the University of Glasgow (Uni-G) ECG analysis program. A 12-lead ECG dataset consisting of 98 confirmed cases of AF and 98 confirmed cases of Sinus rhythm, 49 with PVCs and 49 with PACs, all of which were correctly reported by the Uni-G program and a further 9 cases of confirmed AF which had not been correctly reported by the Uni-G program was used. Each 12-lead ECG was processed to generate two separate ECGs, one with only lead I available and another with only leads I and II available. The sensitivity of reporting AF using a single lead ECG and a 2-lead ECG was identical at 93.8%. However, specificity was greater in the 2-lead ECG at 94.8% compared to 83.6% in the single lead ECG. The results show that a single lead ECG or a 2-lead ECG recording could be effective in screening populations for cardiac arrhythmias.