Key result
A novel AI-algorithm significantly reduced the rate of inconclusive single-lead ECG diagnoses to 1.2% compared to 14-17% with manufacturers' algorithms (p<0.001).
Why the study?
The rate of inconclusive tracings by smart-device manufacturers' algorithms for detecting AF is currently too high to be clinically useful.
Does a novel AI-algorithm improve diagnostic accuracy and reduce inconclusive tracings for atrial fibrillation detection on smart-device single-lead ECGs compared to manufacturers' algorithms in patients presenting to a cardiology service?
Observational (n=206)
No
Does a novel AI-algorithm improve diagnostic accuracy and reduce inconclusive tracings for atrial fibrillation detection on smart-device single-lead ECGs compared to manufacturers' algorithms in patients presenting to a cardiology service?
A novel AI algorithm applied to smart-device single-lead ECGs significantly reduces inconclusive tracings and improves specificity for atrial fibrillation detection compared to native manufacturer algorithms.
Answers are grounded in this paper and cite their sources.
May reduce inconclusive AF diagnoses on smart-device ECGs; leaves open need for prospective validation before practice change.
Background Multiple smart-devices capable of automatically detecting atrial fibrillation (AF) based on single-lead electrocardiograms (SL-ECG) are presently available. The rate of inconclusive tracings by manufacturers' algorithms is currently too high to be clinically useful. Method This is a prospective, observational study enrolling patients presenting to a cardiology service at a tertiary referral centre. We assessed the clinical value of applying a smart-device Artificial Intelligence (AI)-based algorithm for detecting AF from four commercially available smart-devices (AliveCor KardiaMobile, Apple Watch 6, Fitbit Sense, and Samsung Galaxy Watch3). Patients underwent a nearly simultaneous 12-lead ECG and four smart-device SL-ECGs. The novel AI-algorithm (PulseAI, Belfast, United Kingdom) was compared with each manufacturer's algorithm. Results We enrolled 206 patients (31% female, median age 64 years). AF was present in 60 patients (29%). Sensitivity and specificity for the detection of AF by the novel AI-algorithm vs. manufacturer algorithm were: 88% vs. 81% (p=0.34) and 97% vs. 77% (p<0.001) for the AliveCor KardiaMobile, 86% vs. 81% (p=0.45) and 95% vs. 83% (p<0.001) for the Apple Watch 6, 91% vs. 67% (p<0.01) and 94% vs. 82% (p<0.001) for the Fitbit Sense, 86% vs. 82% (p=0.63), and 94% vs. 80% (p<0.001) for the Samsung Galaxy Watch3, respectively. In addition, the proportion of SL-ECGs with an inconclusive diagnosis (1.2%) was significantly lower for all smart-devices using the AI-based algorithm compared to manufacturer's algorithms (14-17%) p<0.001. Conclusion A novel AI-algorithm reduced the rate of inconclusive SL-ECG diagnosis massively while maintaining sensitivity and improving the specificity compared to the manufacturers' algorithms.
No takes yet. Share an insight, caveat, or question.
Weidlich et al. (2023) conducted an observational in Atrial fibrillation (n=206). Novel AI-algorithm (PulseAI) vs. Manufacturers' algorithms (AliveCor KardiaMobile, Apple Watch 6, Fitbit Sense, Samsung Galaxy Watch3) was evaluated on Proportion of SL-ECGs with an inconclusive diagnosis (p=<0.001). A novel AI-algorithm significantly reduced the rate of inconclusive single-lead ECG diagnoses to 1.2% compared to 14-17% with manufacturers' algorithms (p<0.001).
Synapse has enriched 10 related papers. Consider them for comparative context: