Clinical Research Directory

Federated Learning for Point-of-Care Cardiac Ultrasound

This prospective, multicenter study will evaluate a federated machine-learning system designed to analyze focused cardiac point-of-care ultrasound examinations. Federated learning allows participating clinical sites to contribute to model development while keeping raw ultrasound images and directly identifiable patient information within each site's controlled computing environment. Encrypted model updates, rather than patient images, will be transmitted for secure aggregation. The prospective validation cohort will include approximately 3,000 adults undergoing clinically indicated focused cardiac ultrasound. Model performance will be compared with an expert interpretation of a comprehensive transthoracic echocardiogram performed within 24 hours. The primary objective is to determine how accurately the model identifies reduced left ventricular systolic function, defined as a left ventricular ejection fraction below 40%. During the initial validation period, the investigational software will operate in silent mode. Its results will not be displayed to treating clinicians and will not be used to diagnose participants, select treatment, or replace standard clinical interpretation. The study will also evaluate image-quality classification, cardiac-view recognition, performance across clinical sites and ultrasound systems, model calibration, processing time, cybersecurity, privacy resilience, and performance across demographic and clinical subgroups. Long-term monitoring will assess whether model performance changes as clinical populations, ultrasound equipment, acquisition practices, and software environments evolve during the 2026-2037 study period.

COR-INSIGHT: Optimizing Cardiovascular and Cardiopulmonary Outcomes with AI-Driven Multiplexed Indications Using COR ECG Wearable

The COR-INSIGHT trial aims to evaluate the effectiveness of Peerbridge COR advanced ambulatory ECG wearables (COR 1.0 and COR 2.0) in accurately and non-invasively detecting cardiovascular and cardiopulmonary conditions using AI-based software (CardioMIND and CardioQSync). The study devices offer non-invasive, multiplexed, AI-enabled direct-from-ECG detection as a novel alternative to traditional diagnostic methods, including imaging, hemodynamic monitoring systems, catheter-based devices, and biochemical assays. Continuous COR ECG data collected in hospital, outpatient clinic, or home settings will be analyzed to evaluate the predictive accuracy, sensitivity, specificity, and performance of these devices in differentiating between screen-positive and screen-negative subjects. The panel of screened indications encompasses a broad spectrum of clinically relevant cardiovascular, cardiopulmonary, and sleep-related diagnostic parameters, which are critical for advanced patient assessment and management. In the cardiovascular domain, the protocol emphasizes the detection and classification of heart failure, assessment of ejection fraction severity, and identification of myocardial infarction, including pathological Q-waves and STEMI. It further addresses diagnostic markers for arrhythmogenic conditions such as QT interval prolongation, T-wave alternans, and ventricular tachycardia, as well as insights into ischemia, atrial enlargement, ventricular activation time, and heart rate turbulence. Additional parameters, such as heart rate variability, pacing efficacy, electrolyte imbalances, and structural abnormalities, including left ventricular hypertrophy, contribute to comprehensive cardiovascular risk stratification. In the non-invasive cardiopulmonary context, the protocol incorporates metrics like respiratory sinus arrhythmia, cardiac output, stroke volume, and stroke volume variability, providing critical insights into hemodynamic and autonomic function. The inclusion of direct-from-ECG metrics for sleep-related disorders, such as the apnea-hypopnea index, respiratory disturbance index, and oxygen saturation variability, underscores the protocol's utility in addressing the intersection of cardiopulmonary and sleep medicine. This multifaceted approach establishes a robust framework for precision diagnostics and holistic patient management. The COR 1.0 and COR 2.0 wearables provide multi-lead ECG recordings, with COR 2.0 offering extended capabilities for cardiopulmonary metrics and longer battery life (up to 14 days). COR 2.0 supports tri-modal operations: (i) Extended Holter Mode: Outputs Leads II and III, mirroring the functionality of COR 1.0 for broader ECG monitoring applications. (ii) Cardiopulmonary Mode: Adds real-time recording of Lead I, V2, respiratory impedance, and triaxial accelerometer outputs, providing advanced cardiopulmonary insights. (iii) Real-Time Streaming Mode: Streams data directly to mobile devices or computers via Bluetooth Low Energy (BLE), enabling real-time waveform rendering and analysis. The COR 2.0 units are experimental and not yet FDA-cleared. Primary endpoints include sensitivity (true positive rate) \> 80%, specificity (true negative rate) \> 90%, and statistical agreement with reference devices for cardiovascular, cardiopulmonary, and sleep metrics. Secondary endpoints focus on predictive values (PPV and NPV) and overall diagnostic performance. The study employs eight distinct sub-protocols (A through H) to address a variety of cardiovascular, cardiopulmonary, and sleep-related diagnostic goals. These sub-protocols are tailored to specific clinical endpoints, varying in duration (30 minutes to 14 days) and type of data collection. Up to 15,000 participants will be enrolled across multiple sub-protocols. Screening ensures eligibility, and subjects must provide informed consent before participation. Dropouts and non-compliant subjects will be excluded from final analyses.