1120 DIAGNOSTIC ACCURACY OF SMART GADGETS/WEREABLE DEVICES IN ATRIAL FIBRILLATION DETECTION: A METANALYSIS

Abstract Introduction Atrial fibrillation (AF) is the most common sustained arrhythmia and an important risk factor for stroke heart. Recent technology advances have allowed heart rhythm monitoring using smart gadgets/wearable devices which can be used early AF diagnosis. Hypothesis We performed a systemic review meta-analysis to assess accuracy of diagnosis by devices. Methods systematically searched Medline, Embase Cochrane electronic databases up April 15th, 2022 observational studies diagnostic smartphone application, wrist-worn wearables external in detecting AF. calculated area under curve (AUC) summary receiver operating characteristic curves (SROC) pooled sensitivities specificities. Results A total 79 were included enrolling 36903 patients, 66.3% male with average age 68.3±8 years. In overall analysis all devices, AUC was 0.99 (95% CI: 0.98-1.00), sensitivity 95%(95% 94–96%), specificity 96%(95% 96–97%). Wrist-worn had 95%(9% 92–97%), 97%(95% 96–98%)(Figure 1A). Smartphone applications 0.98 0.96-0.99), 96%(9% 94–97%), 93–98%)(Figure 1B). External 93–97%), 95–97%)(Figure 1C). Single-lead ECG 0.98- 1.00), 92–96%), 95–97%). PPG 95–97%), 95–98%). Conclusions single-lead excellent atrial Figure 1. Summary Receiver Operating Characteristic Areas Under The Curve (A), (B) (C).