A Novel Approach Using Remote Speech Analysis In Chronic Ambulatory Heart Failure Patients Allows Early Detection Of Clinical Decompensation Leading To Hospitalization Or Unplanned Iv Diuretics Treatment - Updated New Data

• Published in: Journal of cardiac failure 2022-04, Vol.28 (5), p.S92-S92
• Main Authors: Amir, Offer, Ben-Gal, Tuvia, Weinstein, Jean Marc, Murninkas, Daniel, Lakobishvili, Zaza, Shochat, Michael Kleiner, Dragu, Robert, Ouzan, Elisha, Lotan, Chaim, Edelman, Elazer R., Anker, Stefan D., Shallom, Ilan D., Haviv, Ronit, Pinney, Sean P., Burkhoff, Daniel, Abraham, William T.
• Format: Article
• Language: English
• ISSN: 1071-9164; 1532-8414
• DOI: 10.1016/j.cardfail.2022.03.232

Prevention of acute decompensated heart failure (ADHF) hospitalizations has both medical and economic implications, yet remains an unmet need. The Cordio HearO™ system is a mobile application with cloud-based computing, designed to identify distinct speech measures (SM) in standard audio recordings, which may be indicative of HF clinical status and provide early warning in impending decompensation. Cordio HearO™ speech analysis can identify clinical worsening of heart failure (HF) patients prior to HF events (hospitalization and unplanned IV diuretics treatments). In this ongoing multicenter, study, we recruited 180 NYHA Class II-III, stable stage C chronic HF patients. The patients recorded ∼five sentences (2-5 sec each), in their native language (English, Hebrew, Arabic or Russian), collected daily over a period of 29 months. A total of 63,901 recording days (a total number of available days was 76,313 providing compliance greater than 83%). Of the 24 patient's HF events, 21 (87.5%) (95% CI: 74.3% - 100%) were predicted by the system with a mean of 22 days (min: 3 and max: 31 days) prior to the HF event. The estimated false-positive event rate per year was 2.87 (95% CI: 2.65 - 3.12), one priority every ∼4.2 months (average) per patient. Cordio HearO™ speech analysis novel technology may be a useful tool in remote monitoring of HF patients, providing early warning of impending decompensation resulting in HF events. This approach has the potential to reduce ADHF hospitalizations and improve patient quality of life and economic outcomes.