Abstract
BACKGROUND: The current clinical practice to diagnose atrial fibrillation (AF) requires repeated episodic monitoring and significantly underperform in their ability to detect AF episodes. OBJECTIVE: There is therefore potential for artificial intelligence-based methods to assist in the detection of AF. Better understanding of the optimal parameters for this detection can potentially improve the sensitivity for detecting AF. METHODS: Ten-second, 12-lead electrocardiogram signals were analyzed using complexity algorithms combined with machine learning techniques to predict patients who had a previously detected AF episode but had since returned to normal sinus rhythm. An investigation was performed into the impact of the sampling frequency of the electrocardiogram signal on the accuracy of the machine learning models used. RESULTS: Using a single complexity algorithm showed a peak accuracy of 0.69 when using signals sampled at 125 Hz. In particular, it was noted that improved accuracy occurred when using lead V(6) compared with other available leads. CONCLUSION: Based on these results, there is potential for 12-lead electrocardiogram signals to be recorded at 125 Hz as standard and used in conjunction with complexity analysis to aid in the detection of patients with AF.