ZEUS System (Zio Watch)

The ZEUS System (Zio Watch) is a prescription-based software medical device intended to assist clinicians in managing Atrial Fibrillation by analyzing ECG and PPG data from the Zio Watch using distinct AI algorithms for ECG analysis (ECGDL) and PPG-based AF context (ACE). Its methodology involves AI-driven data processing, machine learning for pattern detection, and subsequent review by Certified Cardiographic Technicians to generate reports for clinical decision support.

The ZEUS System (Zio Watch) is a software as a medical device (SaMD) designed to aid clinicians in the management of Atrial Fibrillation (AF) in adult patients aged 22 years and older who are either diagnosed with or susceptible to developing the condition. Operating as part of the prescription-based Zio Watch Service, the system processes and analyzes electrocardiogram (ECG) and photoplethysmogram (PPG) data collected by the Zio Watch wearable device during ambulatory patient monitoring. Its purpose is to detect and report the presence of AF, providing clinicians with detailed ECG information and contextual data regarding AF episodes, thereby assisting in diagnosis, characterization, and management strategies for non-critical care patients. The system is not intended for use on critical care patients. The core functionality of the ZEUS System (Zio Watch) relies on two distinct artificial intelligence (AI) based algorithms that employ machine learning techniques. The ECG Deep Learned (ECGDL) algorithm analyzes ECG recordings to identify cardiac events such as beats, runs, rhythms, and heart rate. Concurrently, the AF Context Engine (ACE) processes continuously recorded PPG-based data to detect the presence of AF and characterize its occurrence over the monitoring period. Both algorithms were trained using extensive datasets of continuous cardiac recordings. The ACE algorithm underwent validation using multi-day PPG recordings from the Zio Watch compared against ECG-based reference data, while the ECGDL algorithm was validated using ECG data obtained from the Zio Watch. The system compiles these analyses into Zio Watch Transmission Reports, which are subsequently quality-reviewed by Certified Cardiographic Technicians (CCTs) before being made available to clinicians through a secure website.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the AF Context Engine (ACE) algorithm differentiate true Atrial Fibrillation from motion artifacts or other arrhythmias in continuous PPG data, especially during extended ambulatory use?; What are the specific performance metrics (e.g., sensitivity, specificity, positive predictive value, negative predictive value) of the ECGDL and ACE algorithms for AF detection as determined in their respective validation studies (ZWAF Database and Verily Prospective Study)?; How does the ZEUS System (Zio Watch) integrate the discrete ECG recordings with the continuous PPG-derived AF context to provide a unified and actionable report for clinicians, and what is the typical recommended monitoring duration for optimal AF detection?; What specific cybersecurity measures are implemented within the ZEUS System and the Zio Watch Service to ensure the integrity and privacy of patient data throughout its lifecycle, from collection by the Zio Watch to delivery of reports via the secure website?; Given that Certified Cardiographic Technicians (CCTs) review the AI-generated preliminary findings, what is the established protocol for resolving discrepancies between the AI's analysis and the CCT's interpretation, and how is inter-rater reliability among CCTs maintained?