Mobile news

Smartphone voice analysis may identify pulmonary congestion in acute decompensated HF

December 29, 2021

3 min read

The study was supported by Cordio Medical, for which Amir is a paid consultant. Kao reports being an advisor for Codex Health. Ravindra reports no relevant financial disclosures. Please see the study for the other authors’ relevant financial disclosures.

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Automated speech analysis using a smartphone app may be able to detect pulmonary fluid overload in adult patients hospitalized with acute decompensated HF, according to research published in JACC: Heart Failure.

“Most patients with HF present to the hospital with fluid retention, which manifests as worsening dyspnea caused by pulmonary edema,” Offer Amir, MD, director of the Heart Institute at the Hadassah Medical Center in Jerusalem, Israel, and colleagues wrote. “Because pulmonary congestion is not only the predominant contributor to HF hospitalization but also a major predictor of poor postdischarge outcomes, frequent monitoring for pulmonary congestion has been proposed as a means to keep patients well and out of the hospital.”


Source: Adobe Stock

Researchers evaluated whether speech measures were indicative of specific clinical states of pulmonary congestion among patients with acute decompensated HF using a novel speaker verification, speech processing and analysis smart-phone application (HearO, Cordio Medical Or Yehuda). They enrolled 40 adult patients with acute decompensated HF who were asked to record five sentences, repeated three to four times each in their native language (Hebrew, Arabic or Russian).

Recordings were collected upon hospital admission and discharge.

“The premise behind the HearO System is that subtle physiological changes associated with HF decompensation affect the patient’s speech and render them a ‘different person’ (voiceprint),” the researchers wrote. “These changes are much more subtle than those found between different speakers, but are nonetheless detectable using algorithms derived from those used in text-dependent speaker verification.”

Speech measure of pulmonary congestion

Researchers evaluated five unique speech measures that assessed different combinations of characteristics including high temporal resolution, high spectral resolution, autoregressive model for the spectrum, nonlinear amplitude and frequency mapping, symmetric version of a nonlinear spectra ratio and Euclidean distance.

A total of 1,484 recordings were analyzed.

According to the study, recordings taken at hospital discharge were successfully identified as distinctly different from baseline in 94% of cases. In addition, in 87.5% of cases, distinct differences from baseline were detected in all five speech measures.

In a separate analysis, researchers collected an additional 72 voice recordings from nine patients, which were subsequently reanalyzed using the smartphone algorithm in a blinded manner. According to the study, the system isolated the recordings into two distinct unknown sets which, when unblinded, 97.8% of cases successfully corresponded to either admission or discharge recordings.

Differences in recording taken at hospital admission compared with discharge were found for all five speech measures, with the largest observed difference (218%) being found using the second speech measure, which included high temporal resolution, autoregressive model for the spectrum and symmetric version of a nonlinear spectra ratio.

“The current observations provided substantial proof of concept that this novel automated speech processing and analysis approach can reliably identify these differences between two states of pulmonary congestion in patients with HF at the time of hospitalization for acute decompensated HF and following a full course of inpatient treatment,” the researchers wrote. “In this context, this speaker verification-based concept has the potential to serve as a new tool in the in-hospital and the remote armamentarium for assessment of pulmonary congestion in patients with HF.”

‘An important advance’

In a related editorial, Neal G. Ravindra, PhD, postdoctoral fellow in the section of cardiovascular medicine at the Yale School of Medicine, and David P. Kao, MD, associate professor of medicine (cardiology) at the University of Colorado Anschutz Medical Campus, discussed how the study may clear the path for future research into this novel method of detecting pulmonary congestion.

“Active speech analysis as described by Dr Amir et al is an important advance toward expanding the tools available to assess patients with HF,” Ravindra and Kao wrote. “Although nascent, use of commonly available mobile technologies suggests potential for wider use compared with highly invasive strategies requiring dedicated hardware. Extensive development and validation are required before clinical use, but success in a use case such as HearO may pave the way for even more convenient and generalizable strategies.”



This website uses cookies. By continuing to use this site, you accept our use of cookies.