FindTrial
Sign In

Technology that detects danger (congestion) through the voice of a heart failure patient

An Exploratory, Prospective Multicenter Clinical Study investigating an AI-Based Voice Analysis Model for Predicting the Severity of Heart Failure in Patients during Treatment and Controls

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
CRIS
Registry ID
KCT0012156
Enrollment
124
Registered
2026-06-22
Start date
2025-05-30
Completion date
Unknown
Last updated
2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

None listed

Interventions

None listed

Sponsors

Koera University Guro Hospital
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: Patient group: Adults aged 19 and older currently hospitalized with acute heart failure (symptoms such as dyspnea and edema, elevated NT-proBNP, pulmonary congestion on chest radiography, etc.). Control group: Adults aged 19 and older who visit the cardiology department for reasons other than heart failure (hypertension, hyperlipidemia, unspecified chest pain, etc.) and voluntarily provide written consent to participate in this clinical study. Common: Limited to cases where at least two tests to confirm/rule out heart failure (symptoms such as dyspnea and edema, elevated NT-proBNP, pulmonary congestion on chest radiography, etc.) have been performed and the results can be confirmed.

Exclusion criteria

Exclusion criteria: Patients with diseases that may affect the voice, such as pneumonia, sepsis, or lung/vocal cord/laryngeal lesions. Patients with impaired renal function (eGFR <15 mL/min) or those undergoing dialysis. Patients who are hemodynamically unstable or require mechanical ventilation or extracorporeal circulation. Patients for whom voice data collection is difficult due to hearing or speech impairments.

Design outcomes

Primary

MeasureTime frame
Predictive performance (Accuracy, Sensitivity, Specificity, AUC) of the artificial intelligence (AI) model for predicting the severity of heart failure symptoms based on voice data—

Secondary

MeasureTime frame
Clinical indicators and questionnaires representing the severity of heart failure symptoms: Serum biomarker (N-terminal pro b-type natriuretic peptide, NT-proBNP), Bio-impedance analysis (BIA), New York Heart Association (NYHA) Class, Dyspnea Visual Analog Scale (VAS), General condition Visual Analog Scale (VAS), Chest X-ray Congestion Score Index, Quality of Life scale (Kansas City Cardiomyopathy Questionnaire, KCCQ-12 score)—

Countries

Korea, Republic of

Contacts

Public ContactSunki Lee

Koera University Guro Hospital

galiard4@gmail.com+82-2-2626-3183

Outcome results

None listed

Source: CRIS (via WHO ICTRP) · Data processed: Jul 3, 2026