Smartphone speech analysis detects worsening heart failure before an event
October 04, 2022
3 minute read
Abraham W, et al. Late Breaking Clinical Trials Session I. Presented at: Heart Failure Society of America Annual Scientific Meeting; Sept. 30-Oct. 3, 2022; National Harbor, Maryland (hybrid meeting).
Disclosures: Abraham reports receiving a personal fee from Cordio Medical.
NATIONAL HARBOR, Md. — A new speech analysis system was able to detect impending heart failure events in outpatients with congestive heart failure more than 3 weeks in advance, a report reports. speaker.
The system (HearO, Cordio Medical) was not only able to predict HF events with greater sensitivity compared to daily monitoring of weight and oxygen saturation, but also demonstrated fewer average false positives per patient and per year, according to the results of the HearO community study presented at the annual scientific meeting of the Heart Failure Society of America.
“The energy generator of speech are the lungs. We view this analysis technology as primarily reflecting changes in lung fluid content or lung congestion, but it should be noted that hydration can affect the parameters of speech at different levels, including the soft tissues of the vocal tract as well as the vocal cords,” William T. Abraham, MD, a professor of internal medicine and a board-certified physician in advanced cardiology for heart failure and transplantation at Ohio State University Wexner Medical Center, said during a presentation. “The Cordio HearO Community Study is a multi-center, non-interventional, single-arm, open-label clinical study conducted outside of the United States, primarily in Israel, and the goal here was simply to explore diagnostic performance in patients with ambulatory heart failure.”
William T. Abraham
Abraham said the front end of the system is a smartphone app, capable of running on an Android or iOS device, which asks the patient to speak five predefined phrases into the phone every day. From there, it is uploaded to a HIPAA secure cloud-based system where speech analysis and comparison with stable baseline data takes place.
In a previous feasibility study of the system published in JACC: Heart Failureresearchers assessed whether speech measures identified using speaker verification, speech processing, and the smartphone app algorithm would predict clinical states of pulmonary congestion indicative of events of IC in patients with acute decompensated IC.
On admission and discharge from hospital, patients had to record five sentences, repeated three to four times each in their native language.
As Healio previously reported, 94% of cases were identified as markedly different from baseline, and in 87.5% of cases distinct differences from baseline were detected in all five measures of the speech.
For the present analysis, 253 outpatients with congestive heart failure were recruited for the HearO community study (mean age, 69 years; 19% female; 37% with left ventricular ejection fraction
Each patient was asked to make five voice recordings per day for 2 years. Among them, the main language was 72% Hebrew, 19% Russian, 7.5% Arabic and 1.6% English.
Abraham said the speech processing system creates a patient-specific speech pattern while they are in a clinically stable state and compares this pattern in the future with daily speech samples to detect changes the patient has. thought to be related to changes in pulmonary fluid content.
The primary outcome was IC events, defined as hospitalization for IC and treatment with IV diuretics.
Abraham reported that the HearO system could predict about 80% of HF events on average 22.5 days before the events occur.
The researchers observed a false negative rate of about 18%.
Over a total of 186,152 days of analysis, the system’s false positive rate averaged 2.8 alerts per year, translating into one false positive every 4.3 months per patient per year.
The HearO system has demonstrated greater sensitivity for detecting impending HF events compared to current detection standards such as daily weight (80% vs. 35%) and oxygen saturation measurements (80% vs. 25%) and had fewer false positives per patient per year.
Additionally, researchers observed a high rate (70%) of daily voice recording compliance among trial participants.
“Speech analysis technology may be a useful tool in the remote monitoring of patients with heart failure, providing early warning of worsening heart failure, including impending decompensation leading to hospitalization” , Abraham said during the presentation. “Ongoing and future studies will continue to define the role of speech measures in the management of heart failure, and I believe this approach has the potential to reduce [acute decompensated] HF hospitalizations and improve patient quality of life and economic outcomes. But of course, we’ll have to show it in a future prospective, randomized controlled trial.
Abraham added that HearO is currently being evaluated in a new US trial that will test the system’s sensitivity in various dialects, accents and languages representative of the general US population.