A non-invasive method for accurately predicting chronic high blood pressure — otherwise known as hypertension — using just a person's voice has been developed by a team of researchers at Klick Labs.

According to the researchers, 245 study participants recorded their voices every day, up to six times a day for two weeks by speaking into a Klick-developed mobile app. The team found that the device detected high blood pressure with accuracies up to a reported 84% for females and 77% for males.

Source: Klick LabsSource: Klick Labs

To enable this, the app used machine learning to analyze hundreds of vocal biomarkers — such as variability in pitch (fundamental frequency), the patterns in speech energy distribution (Mel-frequency cepstral coefficients) and the sharpness of sound changes (spectral contrast) — which are indiscernible to the human ear.

"By leveraging various classifiers and establishing gender-based predictive models, we discovered a more accessible way to detect hypertension, which we hope will lead to earlier intervention for this widespread global health issue. Hypertension can lead to a number of complications, from heart attacks and kidney problems to dementia," the researchers explained.

The method is detailed in the article, “Machine Learning-Enabled Hypertension Screening Through Acoustical Speech Analysis: Model Development and Validation” which appears in the journal IEEE Access.

To contact the author of this article, email mdonlon@globalspec.com