Clustered Standard Deviation and Its Benefit to Identify Atrial Fibrillation

Abstract

Background: Atrial fibrillation (AF) is a dysfunction of heart atriums shown as irregular heart activity leading to a higher risk of heart failure. Since AF may occur episodically, it is usually diagnosed using ECG Holter recordings. However, the presence of other pathologies and noise makes the automated processing of ECG Holter recordings complicated. Here, we present a new feature to distinguish AF from sinus rhythm as well as from other pathologies: Clustered Standard Deviation (CSTD).Method: QRS complexes are extracted from the ECG signal, and inter-beat intervals (RR) are ordered by their length. Then, RR clusters are found and the mean RR value is computed for each RR cluster. CSTD is computed using a formula for standard deviation using cluster-specific mean values instead of a global mean.Results: CSTD was evaluated for 7,254 ECG segments from a private dataset (MDT company, Brno, Czechia), 60 seconds length, 1-lead, 250 Hz sampling frequency. CSTD showed high values for AF while remaining low for other pathologies and sinus rhythm. CSTD between AF and other classes showed AUC 0.95. For comparison, a standard deviation of RR intervals leads to AUC 0.65 due to its sensitivity to other pathologies. Test on public MIT-AFDB dataset shown AUC and AUPRC 0.98 and 0.97, respectively.