Continuous Wearable Monitoring Analytics Predict Heart Failure Hospitalization: The LINK-HF Multicenter Study.

Area of Science:

• Cardiology
• Biomedical Engineering
• Health Informatics

Background:

• Implantable cardiac sensors show potential for reducing heart failure (HF) rehospitalizations.
• The effectiveness of noninvasive remote monitoring for predicting HF rehospitalization remains undetermined.

Purpose of the Study:

• To assess the accuracy of noninvasive remote monitoring in predicting HF rehospitalization.
• To evaluate a personalized analytical platform using continuous data streams for HF exacerbation prediction.

Main Methods:

• The LINK-HF study enrolled 100 patients post-HF admission, monitored with a chest-worn multisensor patch for up to 3 months.
• Physiological data were continuously uploaded via smartphone to a cloud platform for analysis.
• Machine learning developed a prognostic algorithm to detect HF exacerbation, with clinical events adjudicated.

Main Results:

• The platform achieved 76%–88% sensitivity and 85% specificity in detecting precursors to HF hospitalization.
• A median of 6.5 days' warning was provided between the initial alert and readmission.
• The system derived personalized baseline physiological models to identify deviations.

Conclusions:

• Multivariate physiological telemetry from wearable sensors enables accurate early detection of impending HF rehospitalization.
• This noninvasive approach demonstrates predictive accuracy comparable to implanted devices.
• Further testing is recommended to confirm the clinical efficacy and generalizability of this low-cost monitoring method.