Novel Speech Recognizing App may help Predict HF by recognising fluid buildup in lungs: Study
Heart failure (HF) is a progressive condition that affects approximately 26 million people worldwide. Most patients with HF present to the hospital with fluid retention, which manifests as worsening dyspnea caused by pulmonary edema. In a recent study, researchers have developed a novel mobile app, HearO that detects changes in speech which predicts clinical congestion in patients with HF. The study findings were published in the journal JACC: Heart Failure on December 8, 2021.
Recent advances in speech, voice, and sound analysis enabled the identification of speech features of clinical significance. Various speech processing algorithms have been developed to use such features in screening for depression, pneumonia-asthma, coronary artery disease, and an autism spectrum disorder. Pulmonary edema is the main cause of heart failure (HF)−related hospitalizations. It is an important predictor of poor prognosis after discharge. Frequent monitoring is often recommended, but signs of decompensation are overlooked. Recently, Dr Offer Amir and his team conducted a study to assess the performance of an automated speech analysis technology in detecting pulmonary fluid overload in patients with acute decompensated heart failure (ADHF). They distinguishing between ADHF on admission ("wet") versus at discharge ("dry") by using a novel speaker verification, speech processing, and analysis technology, implemented within a proprietary Smartphone application (app) (HearO Cordio Medical Ltd, Or Yehuda, Israel).
In this observational open-label study, the researchers included a total of 40 patients with ADHF and recorded 5 sentences, in 1 of 3 languages, using HearO app, upon admission (wet) to and discharge (dry) from the hospital. HerO app is a proprietary speech processing application that was used to record and digitize the patients' speech. The researchers uploaded the speech files to the server, where they were stored and analyzed. Recordings were analyzed for 5 distinct speech measures (SMs), each a distinct time, frequency resolution, and linear versus perceptual (ear) model. They further calculated the mean change from baseline SMs.
Article Source : JACC Heart Failure
