AI-Based Monitoring System for Chronic Heart Failure With Advanced Wearable and Mini-Invasive Devices

Not yet recruiting

• Conditions: Chronic Heart Failure; Cardiovascular Diseases; Heart Failure With Reduced Ejection Fraction (HFrEF); Heart Failure With Preserved Ejection Fraction (HFPEF); Congestive Heart Failure Chronic

• Registration Number: NCT06909682
• Lead Sponsor: University of Salerno

Brief Summary

The goal of this observational, multicenter study is to evaluate whether AI-driven remote monitoring using a mini-invasive wearable device can improve clinical outcomes in adult patients (≥18 years) with chronic heart failure (CHF).

The main questions it aims to answer are:

* Can continuous remote monitoring reduce hospital admissions (emergency visits and hospitalizations) by 20% compared to standard care?

* Does wearable-based remote monitoring improve functional, biochemical, and instrumental parameters in CHF patients? Researchers will compare patients using the wearable device (intervention group) to those receiving standard clinical follow-up (control group) to assess whether AI-driven monitoring leads to fewer hospitalizations, better disease management, and improved quality of life.

Participants will:

* Wear the EmbracePlus (Empatica Inc.) device continuously for six months (intervention group only).

* Have their biometric data (SpO₂, HRV, EDA, respiratory rate, temperature, sleep quality) monitored remotely.

* Receive automated alerts and teleconsultations if abnormal physiological changes are detected.

* Attend scheduled follow-up visits (remote and in-person) for clinical evaluation and treatment adjustments.

The study aims to provide real-world evidence on whether integrating wearable health technology with AI analytics can enhance CHF management and improve patient outcomes.

Detailed Description

Chronic Heart Failure (CHF) is a multifactorial syndrome characterized by high rates of hospitalization, morbidity, and mortality. Despite advances in pharmacological and device-based therapies, early identification of clinical deterioration remains a major challenge. Traditional follow-up models, based primarily on intermittent in-person evaluations, are often inadequate in capturing subclinical changes that precede acute decompensation.

The SMART-CARE (System of Monitoring and Analysis based on Artificial Intelligence for Chronic Heart Failure Patients with Mini-Invasive and Wearable Medical Devices) study aims to assess whether continuous remote monitoring using a CE (Conformité Européenne)-certified wearable device (EmbracePlus by Empatica Inc.) integrated with AI (Artificial Intelligence) analytics can improve the management of CHF patients. The study adopts a prospective, multicenter, observational design with two parallel cohorts: patients managed with standard care versus patients equipped with the wearable device for six months.

The wearable device captures a range of physiological signals-including peripheral capillary oxygen saturation (SpO₂), heart rate variability (HRV), electrodermal activity (EDA), skin conductance level (SCL), respiratory rate, peripheral skin temperature, pulse rate, fatigue detection, and sleep metrics via actigraphy-and transmits them in real time to a centralized digital platform. AI algorithms analyze these data continuously, triggering alerts in the event of abnormal trends. When alerts are generated, patients undergo teleconsultation, with possible treatment adjustments or in-person follow-up as clinically indicated.

The study is designed to generate real-world evidence on whether AI-enhanced monitoring can reduce unplanned hospital admissions by at least 20% over a six-month follow-up, compared to standard care. Secondary endpoints include improvements in cardiac function (evaluated through echocardiographic parameters), neurohormonal biomarkers such as B-type Natriuretic Peptide (BNP) and Atrial Natriuretic Peptide (ANP), exercise tolerance assessed by the Six-Minute Walk Test (6MWT), quality of life measured by the Kansas City Cardiomyopathy Questionnaire (KCCQ), and incidence of therapy-related adverse events (e.g., hypotension, bradyarrhythmias).

In addition to evaluating clinical efficacy, the study supports the development of a predictive multimarker model. Data collected through the SMART-CARE platform-including clinical history, biochemical markers, imaging data, and continuous sensor-derived variables-will be used by collaborating academic centers to train AI algorithms capable of forecasting CHF progression and tailoring individualized interventions.

All data are pseudonymized in compliance with the General Data Protection Regulation (GDPR, Regulation EU 2016/679). The study does not interfere with ongoing medical treatments and adheres to Good Clinical Practice (GCP) and the ethical principles of the Declaration of Helsinki. Patients provide written informed consent prior to enrollment.

The SMART-CARE initiative reflects a broader goal: integrating telemedicine, wearable health technology, and AI-based predictive modeling into a seamless care pathway that promotes proactive CHF management and enables personalized, data-driven therapeutic decisions.

Study Timeline

• Status Verified: 2025-03
• Study First Submitted: 2025-03-14
• Study First Submitted QC: 2025-04-01
• Last Update Submitted: 2025-04-01
• Study First Posted: 2025-04-03
• Last Update Posted: 2025-04-03
• Study Start: 2025-08-01
• Primary Completion: 2026-08-01
• Study Completion: 2027-02-02

Recruitment & Eligibility

• Status: NOT_YET_RECRUITING
• Sex: All
• Target Recruitment: 205

Inclusion Criteria

Not provided

Exclusion Criteria

Not provided

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Change in Hospital Admissions with AI-Based Remote Monitoring	6 months from participant enrollment.	The study aims to determine whether AI-based remote monitoring using a wearable device leads to a 20% reduction in hospital admissions (including emergency department visits and hospitalizations) compared to standard clinical follow-up in patients with chronic heart failure (CHF). The intervention group will use a mini-invasive wearable device for continuous physiological monitoring, while the control group will receive standard CHF management without remote monitoring. Hospital admission rates will be analyzed to assess the effectiveness of early AI-driven detection and intervention.

Secondary Outcome Measures

Name	Time	Method
Change in Quality of Life	Baseline, 3 months, and 6 months	Quality of life (QoL) will be measured using the Kansas City Cardiomyopathy Questionnaire (KCCQ) Overall Summary Score, a validated instrument specifically designed to assess symptom burden, functional status, social limitations, and quality of life in patients with chronic heart failure (CHF). The score ranges from 0 to 100, where higher scores indicate better health status and quality of life. The study will evaluate whether patients in the AI-based remote monitoring group report higher KCCQ scores compared to the control group.; Unit of Measure: KCCQ score (0-100 scale) Time Frame: Baseline, 3 months, and 6 months Interpretation: Higher scores indicate better outcomes.
Adverse Effects of CHF Therapy	6 months	The study will analyze whether continuous AI-driven monitoring helps in reducing adverse effects related to CHF treatments, such as: Hypotension (low blood pressure episodes due to overuse of diuretics or vasodilators) Bradyarrhythmias (slow heart rate linked to beta-blockers or other heart failure medications) By detecting early physiological changes, the wearable device may enable timely adjustments in medication dosages, reducing complications and therapy-related hospitalizations.
Change in Biochemical Parameters	3 and 6 months from participant enrollment	Biochemical Parameters: Change in Serum Chloride Levels (millimoles per liter)
Change in Functional ECG-Derived Parameters	3 and 6 months from participant enrollment	ECG-Derived Parameters: Change in Respiratory Rate (breaths per minute)
Change in Functional Echocardiographic derived Parameters	3 and 6 months from participant enrollment	Echocardiographic Parameters: Change in Left Ventricular End-Systolic Volume (milliliters)