Electrocardiographic Diagnostic Performance of the Apple Watch Augmented With an Artificial Intelligence Algorithm

Completed

• Conditions: Premature Supraventricular Beat; Tachycardia; Premature Ventricular Contraction; Bradycardia; Atrial Fibrillation

• Registration Number: NCT05167058
• Lead Sponsor: Cardiologs Technologies

Brief Summary

The SWAF study will compare the performance of a smartwatch combined with Cardiologs Platform algorithm in the detection of Atrial Fibrillation and other arrhythmias with that measured on a manually read 12-lead ECG in subjects hospitalized for cardioversion or AF ablation.

Detailed Description

The SWAF study is a prospective, non-significant risk, non-randomized, multicentric, open, comparative, confirmatory study.

Under subject consent, subjects hospitalized for cardioversion or AF ablation will have a smartwatch ECG recording done simultaneously with 12-lead ECG measurement right before the intervention. If a subject is found in Normal Sinus Rhythm he/she will be discharged otherwise the patient will undergo cardioversion and will have simultaneous recordings done a second time after the intervention. All the measurements will be done in accordance with the existing subject monitoring protocol.

Study Timeline

• Study First Submitted: 2021-12-08
• Study First Submitted QC: 2021-12-08
• Study First Posted: 2021-12-22
• Study Start: 2022-04-21
• Primary Completion: 2023-01-20
• Study Completion: 2023-03-31
• Status Verified: 2023-10
• Last Update Submitted: 2023-10-02
• Last Update Posted: 2023-10-03

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 220

Inclusion Criteria

• Subjects over 18, able and willing to participate in the study
• Subjects who are admitted to the hospital for a cardioversion or AF ablation procedure.
• Subjects having read the patient information letter and provided his/her consent to participate in writing, by dating and signing the informed consent prior to any trial-related procedure being conducted

Exclusion Criteria

• Subjects with paced rhythm or implanted electronic devices
• Pregnant or breast-feeding subjects

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting AF (Atrial Fibrillation or Flutter) as identified by the physician on the 12-lead ECG	Readings taken simultaneously right before and right after the cardioversion according to the monitoring protocol at the hospital	Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting AF (Atrial Fibrillation or Flutter) as identified by the physician on the 12-lead ECG in the independent annotation center, providing the ground truth from the 12-lead ECG

Secondary Outcome Measures

Name	Time	Method
Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting AF as identified by the physician on the smartwatch ECG	Readings taken simultaneously right before and right after the cardioversion according to the monitoring protocol at the hospital	Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting AF as identified by the physician on the smartwatch ECG in the independent annotation center, providing the ground truth from the smartwatch ECG.
Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Sinus Rhythm as identified by the physician on the smartwatch ECG	Readings taken simultaneously right before and right after the cardioversion according to the monitoring protocol at the hospital	Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Sinus Rhythm as identified by the physician on the smartwatch ECG in the independent annotation center, providing the ground truth from the smartwatch ECG.
Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Premature Supraventricular Complexes as identified by the physician on the smartwatch ECG	Readings taken simultaneously right before and right after the cardioversion according to the monitoring protocol at the hospital	Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Premature Supraventricular Complexes as identified by the physician on the smartwatch ECG in the independent annotation center, providing the ground truth from the smartwatch ECG.
Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Premature Ventricular Complexes as identified by the physician on the smartwatch ECG	Readings taken simultaneously right before and right after the cardioversion according to the monitoring protocol at the hospital	Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Premature Ventricular Complexes as identified by the physician on the smartwatch ECG in the independent annotation center, providing the ground truth from the smartwatch ECG.
Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Tachycardia as identified by the physician on the smartwatch ECG	Readings taken simultaneously right before and right after the cardioversion according to the monitoring protocol at the hospital	Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Tachycardia as identified by the physician on the smartwatch ECG in the independent annotation center, providing the ground truth from the smartwatch ECG.
Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Bradycardia as identified by the physician on the smartwatch ECG	Readings taken simultaneously right before and right after the cardioversion according to the monitoring protocol at the hospital	Evaluation of the performance of smartwatch ECG interpreted by Cardiologs Artificial Intelligence in detecting Bradycardia as identified by the physician on the smartwatch ECG in the independent annotation center, providing the ground truth from the smartwatch ECG.
Assessment of the proportion of smartwatch ECGs identified as inconclusive by Cardiologs Artificial Intelligence and by the physician	Readings taken simultaneously right before and right after the cardioversion according to the monitoring protocol at the hospital	Assessment of the proportion of smartwatch ECGs identified as inconclusive by Cardiologs Artificial Intelligence and by the physician. Inconclusive may mean that there may have been too much artefact or noise to acquire a good signal or that the rhythm is unclassifiable or contains other abnormal rhythm.

Trial Locations

Locations (3)

The Valley Hospital; 🇺🇸 Ridgewood, New Jersey, United States

Hackensack Meridian School of Medicine; 🇺🇸 Hackensack, New Jersey, United States

Columbia University Medical Center/ NewYork Presbyterian Hospital; 🇺🇸 New York, New York, United States