Wearables and Artificial Intelligence in Advanced Heart Failure Care

N/ANot Yet RecruitingNCT07051356

UMC Utrecht200 enrolled

Overview

The goal of this observational study is to evaluate whether AI-based analyses of wearable sensor data can identify early signs of deterioration leading to hospitalization in patients with advanced heart failure. The main questions it aims to answer are: * Can AI-driven analysis of wearable data detect physiological or behavioral changes associated with impending hospital admissions? * Does wearable-based remote monitoring influence daily exercise duration in patients with advanced heart failure. * Is wearable-based remote monitoring usable and acceptable for patients with advanced heart failure in a real-world setting? Participants will wear a wrist-worn (Fitbit) device continuously for one year and will use an eHealth app to answer question about their symptoms. Participant's physical activity, heart rate, heart rate variability, respiratory rate, sleep quality, and symptomatic status will be monitored remotely.

Advanced heart failure (HF) is characterized by persistent and progressive symptoms despite optimal, guideline-directed medical therapy. Although improvements in care have been achieved, mortality remains high, and recurrent hospitalizations continue to significantly impact patients' morbidity and quality of life. Timely recognition of early signs of clinical deterioration remains a challenge. Innovative approaches that enable early identification of patients at increased risk of readmission may support proactive interventions and help reduce the need for hospitalization. In the WAI-HF study, we will investigate whether AI-driven analysis wearable data can identify changes that precede hospital admission in patients with advanced heart failure. The wrist-worn device measures several physiological parameters including heart rate, heart rate variability, respiratory rate, skin temperature, 1-lead electrocardiogram, and sleep quality. Data collected in the remote monitoring including continuous data derived from the wearable device and symptomatic data collected in the eHealth app, will be used to develop a predictive model. The study will be conducted according to the principles of the Declaration of Helsinki (64th WMA General Assembly, Fortaleza, Brazil, October 2013), to 'gedragscode gezondheidsonderzoek', and in accordance with the EU GDPR (General Data Protection Regulation).

Study Type

OBSERVATIONAL

Enrollment

200

Eligibility

Sex: ALLMin age: 18 Years

Medical Language ↔ Plain English

Inclusion Criteria: * \>18 years. * Diagnosis of advanced heart failure, including at least one of the following major criteria. * LVAD implanted * Included on the waiting list for Heart transplant * Meeting the European Society of CArdiology criteria for advanced HF: * Severe and persistent symptoms of heart failure \[NYHA class III or IV\]. * Severe cardiac dysfunction: according to ESC guidelines definition * ≥ 1 unplanned visit or hospitalization in the last 12 months requiring IV treatment. * Have access to a mobile phone or tablet with an operating system iSO 15 or Android 9 (or posterior versions of these systems). Exclusion Criteria: * Impossibility to provide inform consent. * Impossibility to self-report data due to physical or mental disability.

Outcomes

Primary Outcomes

Algorithm Performance Metrics

Algorithm's performance to detect imminent admission in patients with advanced HF will be measured by means of the following parameters: Accuracy, sensitivity, specificity, negative predictive value, positive predictive value, area under the ROC curve.

Time frame: From enrollment to the end of the monitoring period at 1 year.

Secondary Outcomes

Change in daily exercise duration

Daily exercise duration will be measured using the wrist worn-device. Exercise duration will be measured in minutes of activity per day and step count per day.

Time frame: From baseline to the end of the monitoring period at 1 year.

Perceived usability

Perceived usability will be assessed by means of The System Usability Scale (SUS). SUS is a widely used, validated tool for assessing the usability of a system, product, or technology. SUS is a 10-item questionnaire with statements about the system being evaluated. Each item is rated on a 5-point Likert scale (from Strongly Disagree \[1\] to Strongly Agree \[5\]). The single score ranges from 0 to 100 where higher scores indicate better usability.

Time frame: At 1-year