Scientists develop stunning AI to predict blood pressure from voice

Researchers at Klick Labs have made a groundbreaking discovery in health monitoring. They’ve developed a new, non-invasive technique that predicts chronic high blood pressure (hypertension) using just a person’s voice.

This innovative method offers an exciting glimpse into the future of early hypertension detection. By harnessing vocal biomarkers, Klick Labs aims to improve health outcomes with a tool that’s both accessible and effective.

How it works: Voice analysis for better health

In their study, Klick Labs involved 245 participants who recorded their voices up to six times a day for two weeks. They used a mobile app designed by Klick Labs, which analyzes these voice recordings to detect high blood pressure .

The app’s machine learning algorithms examine subtle vocal features, such as pitch variability, speech energy patterns, and sound changes, to predict hypertension with notable accuracy. For women, the app achieved up to 84 percent accuracy, while for men, it reached 77 percent.

Yan Fossat, senior vice president of Klick Labs and principal investigator of the study, emphasized the significance of their findings.

“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,” Fossat said.

Addressing the “silent killer”

Hypertension, often dubbed the “silent killer” by the World Health Organization (WHO), affects over 25 percent of the global population. Unfortunately, many are unaware of their condition, and the majority of those diagnosed live in low- or middle-income countries.

Traditional methods of blood pressure measurement, such as using an arm cuff or automatic devices, can be costly and require technical expertise. This makes it challenging for people in underserved areas to access these crucial screenings.

Klick Labs’ new voice-based technique presents a promising solution. It offers a way to screen for hypertension that is both easier and more accessible than traditional methods. This study marks Klick Labs’ first major step into using voice technology for health conditions beyond diabetes.

Their previous research demonstrated that voice analysis combined with AI could effectively screen for Type 2 diabetes, as published in Mayo Clinic Proceedings: Digital Health in October 2023. Additionally, a recent study in Scientific Reports confirmed the relationship between blood glucose levels and voice pitch.

Jaycee Kaufman, a research scientist at Klick Labs and co-author of the study, highlighted the potential impact of this technology. “Voice technology has the potential to exponentially transform healthcare, making it more accessible and affordable, especially for large, underserved populations,” Kaufman stated.

“Our ongoing research increasingly demonstrates the significant promise of vocal biomarkers in detecting hypertension, diabetes, and a growing list of other health conditions.”

The study was recently detailed in the peer-reviewed journal IEEE Access .