Trends in voice characteristics in patients with heart failure (VENTURE) in Switzerland: Protocol for a longitudinal observational pilot study
Journal
PLOS ONE
ISSN
1932-6203
Type
journal article
Date Issued
2023-04-05
Author(s)
Fan Wu
;
Matthias Nägele
;
David Cleres
;
Thomas Haider
;
;
Frank Ruschitzka
;
Andreas Flammer
;
Filipe Barata
Editor(s)
Hugh Cowley
Abstract
<jats:sec id="sec001">
<jats:title>Introduction</jats:title>
<jats:p>Heart Failure (HF) is a major health and economic issue worldwide. HF-related expenses are largely driven by hospital admissions and re-admissions, many of which are potentially preventable. Current self-management programs, however, have failed to reduce hospital admissions. This may be explained by their low predictive power for decompensation and high adherence requirements. Slight alterations in the voice profile may allow to detect decompensation in HF patients at an earlier stage and reduce hospitalizations. This pilot study investigates the potential of voice as a digital biomarker to predict health status deterioration in HF patients.</jats:p>
</jats:sec>
<jats:sec id="sec002">
<jats:title>Methods and analysis</jats:title>
<jats:p>In a two-month longitudinal observational study, we collect voice samples and HF-related quality-of-life questionnaires from 35 stable HF patients. Patients use our developed study application installed on a tablet at home during the study period. From the collected data, we use signal processing to extract voice characteristics from the audio samples and associate them with the answers to the questionnaire data. The primary outcome will be the correlation between voice characteristics and HF-related quality-of-life health status.</jats:p>
</jats:sec>
<jats:sec id="sec003">
<jats:title>Ethics and dissemination</jats:title>
<jats:p>The study was reviewed and approved by the Cantonal Ethics Committee Zurich (BASEC ID:2022-00912). Results will be published in medical and technical peer-reviewed journals.</jats:p>
</jats:sec>
<jats:title>Introduction</jats:title>
<jats:p>Heart Failure (HF) is a major health and economic issue worldwide. HF-related expenses are largely driven by hospital admissions and re-admissions, many of which are potentially preventable. Current self-management programs, however, have failed to reduce hospital admissions. This may be explained by their low predictive power for decompensation and high adherence requirements. Slight alterations in the voice profile may allow to detect decompensation in HF patients at an earlier stage and reduce hospitalizations. This pilot study investigates the potential of voice as a digital biomarker to predict health status deterioration in HF patients.</jats:p>
</jats:sec>
<jats:sec id="sec002">
<jats:title>Methods and analysis</jats:title>
<jats:p>In a two-month longitudinal observational study, we collect voice samples and HF-related quality-of-life questionnaires from 35 stable HF patients. Patients use our developed study application installed on a tablet at home during the study period. From the collected data, we use signal processing to extract voice characteristics from the audio samples and associate them with the answers to the questionnaire data. The primary outcome will be the correlation between voice characteristics and HF-related quality-of-life health status.</jats:p>
</jats:sec>
<jats:sec id="sec003">
<jats:title>Ethics and dissemination</jats:title>
<jats:p>The study was reviewed and approved by the Cantonal Ethics Committee Zurich (BASEC ID:2022-00912). Results will be published in medical and technical peer-reviewed journals.</jats:p>
</jats:sec>
Language
English (United States)
Volume
18
Number
4