PO-04-177 WEARABLE TECHNOLOGY FOR ATRIAL FIBRILLATION DETECTION: A COMPARISON OF THE APPLE WATCH AND THE CART-I RING IN PATIENT WITH CARDIOVASCULAR DISEASE

Wearable technology is increasingly used to identify episodes of atrial fibrillation (AF). However, most validation studies using single-lead ECGs only included patients with either sinus rhythm or AF and, therefore, likely overestimate the sensitivity and specificity detection algorithms. We sought compare diagnostic accuracy Apple Watch (APW) series 6 CART-I ring (C-ring) algorithms physician-interpreted in known cardiovascular disease. In this single-centre prospective study, individuals disease underwent simultaneous 12-lead ECG recordings a 30-second from wearable device. The order devices was randomly assigned. All were independently adjudicated by two cardiologists, any disagreements resolved third cardiologist. devices’ algorithm labelled as "AF", "Not AF" "Unclassified". determined calculating for each device compared Fisher’s exact test McNemar test. Four hundred consecutive enrolled but 7 excluded due uninterpretable ECGs. Baseline demographics are shown Table 1. A total 1573 analysed. Forty-two percent AF. Level agreement between adjudicators high (89.8% APW vs 91.1% C-ring, p= 0.54).The had more Unclassified than C-ring (78 vs. 2, respectively). If all included, C- ring's significantly higher APW, at 86.6% (95% CI:80.6%-91.1%) 67.4% (95%CI:60.0%-74.1%), respectively. remain even after excluding unclassified ECGs, not significant (mean difference 7.52%, p=0.09). Physician interpretation sl-ECGs led increase specificity, 95.8% CI:91.6%-97.9%) 90.8 % CI:86.2%-93.9%), respectively (Figure). contrast, physician C-ringsl-ECG resulted (94.6% [95% CI:90.0%-97.1%]) marginal decrease (89.4% CI:84.8%-92.8%]). There no statistically sl-ECGs. Automated diagnosis suboptimal, number false-positives ECGs; however, both devices.