Artificial Intelligence-Based Disease Management in the Vulnerable Period of Heart Failure

Overview

The study aims to evaluate the effect of AI-based discharge training after acute decompensation of heart failure on the patient's quality of life and to examine the relationship between changes in voice and speech characteristics of patients and changes in hospitalization, discharge, and early post-discharge clinical status.

Heart failure (HF) is a progressive disease with a fluctuating course. From time to time, patients' symptoms and signs worsen enough to require hospitalization, and hospitalization occurs with acute decompensated HF. Acute HF decompensations are periods that worsen the prognosis of the patient. On the other hand, patients discharged after an acute decompensation have the highest risk during the first months after discharge. Patients get trained about lifestyle changes and medication management during the discharge period. Recent guidelines suggest that patients in the early post-discharge period be called for virtual/telephone or face-to-face control visits at short intervals(3rd, 7th, 14th, and 28th days ), then at 3-6 monthly intervals according to NYHA class. In traditional cardiovascular practice, patients are called for outpatient control visit in the first month after discharge. However, the processes after the pandemic kept patients away from visiting the hospital at the frequency recommended in the American and European guidelines and caused them to stay at home not even following their routine visit schedules. Moreover, the risks and benefits of virtual visits in terms of patient prognosis are not well established. This study aims to investigate relationships between routinely recorded findings, symptoms, and vocal biomarkers of heart failure patients in the hospitalization and in the post-discharge period and to investigate the effect of post-discharge education on patient-reported outcomes and re-hospitalization.

Conditions

Eligibility

Locations (1)

Outcomes