Problem
Research questions and friction points this paper is trying to address.
heart failure
digital biomarkers
voice analysis
health deterioration prediction
chronic disease monitoring
Vocal Prognostic Digital Biomarkers in Monitoring Chronic Heart Failure: A Longitudinal Observational Study
🤖 AI Summary
This study addresses the critical gap in high-precision, low-burden early warning tools for deterioration in chronic heart failure. Through a two-month longitudinal at-home monitoring protocol, daily concurrent voice and routine physiological data were collected from patients. By integrating temporal modeling with interpretable machine learning, the research identifies, for the first time during clinical stability, predictive vocal digital biomarkers—such as delayed energy shift and reduced phonation rate—that overcome the limitations of conventional metrics like weight. The resulting model achieves a sensitivity of 0.826 and specificity of 0.782 in forecasting next-day health status changes, significantly outperforming standard-of-care indicators. This approach enables seamless, highly adherent early detection without imposing additional burden on patients.
📝 Abstract
Objective: This study aimed to evaluate which voice features can predict health deterioration in patients with chronic HF.
Background: Heart failure (HF) is a chronic condition with progressive deterioration and acute decompensations, often requiring hospitalization and imposing substantial healthcare and economic burdens. Current standard-of-care (SoC) home monitoring, such as weight tracking, lacks predictive accuracy and requires high patient engagement. Voice is a promising non-invasive biomarker, though prior studies have mainly focused on acute HF stages.
Methods: In a 2-month longitudinal study, 32 patients with HF collected daily voice recordings and SoC measures of weight and blood pressure at home, with biweekly questionnaires for health status. Acoustic analysis generated detailed vowel and speech features. Time-series features were extracted from aggregated lookback windows (e.g., 7 days) to predict next-day health status. Explainable machine learning with nested cross-validation identified top vocal biomarkers, and a case study illustrated model application.
Results: A total of 21,863 recordings were analyzed. Acoustic vowel features showed strong correlations with health status. Time-series voice features within the lookback window outperformed corresponding standard care measures, achieving peak sensitivity and specificity of 0.826 and 0.782 versus 0.783 and 0.567 for SoC metrics. Key prognostic voice features identifying deterioration included delayed energy shift, low energy variability, and higher shimmer variability in vowels, along with reduced speaking and articulation rate, lower phonation ratio, decreased voice quality, and increased formant variability in speech.
Conclusion: Voice-based monitoring offers a non-invasive approach to detect early health changes in chronic HF, supporting proactive and personalized care.
Innovation
Methods, ideas, or system contributions that make the work stand out.
vocal biomarkers
chronic heart failure
explainable machine learning
time-series voice features
digital health monitoring
🔎 Similar Papers
No similar papers found.
