Hypertension, often referred to as the “silent killer,” remains a major global health challenge. Affecting over 35% of the population worldwide, this condition often goes undetected until complications arise.
Despite advancements in blood pressure monitoring devices, conventional methods remain inaccessible to many, especially in underserved regions. This has sparked innovative approaches to screening, including a groundbreaking study that explores voice analysis as a non-invasive diagnostic tool for hypertension.
Researchers at Klick Labs have pioneered a technique to predict chronic high blood pressure using just a person’s voice.
Their study, recently published in the peer-reviewed journal IEEE Access, showcases the potential of vocal biomarkers in detecting hypertension. By analyzing subtle patterns in speech, the method promises to make health screenings more accessible and affordable.
The study involved 245 participants, who recorded their voices up to six times daily for two weeks using a proprietary mobile app. This app, developed by Klick Labs, employs machine learning to analyze hundreds of vocal biomarkers undetectable to the human ear. These include variations in pitch, energy distribution patterns, and changes in sound sharpness.
The researchers established two thresholds for defining hypertension: systolic blood pressure (SBP) of ≥135 mmHg or diastolic blood pressure (DBP) of ≥85 mmHg; and SBP of ≥140 mmHg or DBP of ≥90 mmHg.
Predictive models were tailored by gender, achieving up to 84% accuracy for females and 77% for males under the first threshold. For the stricter second threshold, accuracy was 63% for females and 86% for males.
“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,” said Yan Fossat, senior vice president of Klick Labs and principal investigator of the study. “Hypertension can lead to a number of complications, from heart attacks and kidney problems to dementia.”
Hypertension is a leading cause of heart disease, stroke, and kidney failure, making it a critical public health concern. The World Health Organization (WHO) estimates that over 50% of individuals with hypertension remain unaware of their condition. Alarmingly, more than 75% of diagnosed cases occur in low- and middle-income countries, where access to healthcare is limited.
Related Stories
Traditional blood pressure measurement techniques, such as using arm cuffs or automatic devices, require technical expertise and specialized equipment. These limitations hinder widespread screening, particularly in remote or underserved areas. The innovative voice-analysis method developed by Klick Labs addresses this gap by offering a cost-effective and portable solution.
Voice-based detection relies on identifying patterns within speech that correlate with physiological states. The mobile app designed for this study analyzes biomarkers like fundamental frequency (pitch), Mel-frequency cepstral coefficients (speech energy distribution), and spectral contrast (sharpness of sound changes). These features reflect subtle changes in the body’s cardiovascular system, offering a window into blood pressure levels.
Machine learning algorithms trained on these biomarkers enable the app to classify individuals as hypertensive or normotensive with remarkable accuracy. The study’s leave-one-subject-out validation approach ensured robust testing of the models, minimizing bias and enhancing reliability.
“Voice technology has the potential to exponentially transform healthcare, making it more accessible and affordable, especially for large, underserved populations,” said Jaycee Kaufman, a research scientist at Klick Labs and co-author of the study. “Our ongoing research increasingly demonstrates the significant promise of vocal biomarkers in detecting hypertension, diabetes, and a growing list of other health conditions.”
This research marks Klick Labs’ first foray into using voice technology to identify hypertension, building on their earlier work in diabetes detection. In October 2023, the team published findings in Mayo Clinic Proceedings: Digital Health, demonstrating that voice analysis combined with AI could accurately screen for Type 2 diabetes. Another study, featured in Scientific Reports, confirmed a link between blood glucose levels and voice pitch.
By collaborating with hospitals, academic institutions, and public health authorities worldwide, Klick Labs aims to refine its AI algorithms for broader applications. From early detection of chronic diseases to continuous health monitoring, vocal biomarkers could revolutionize the way we approach preventive care.
“Hypertension is just the beginning,” said Fossat. “As our understanding of vocal biomarkers deepens, we envision a future where a simple voice recording could provide insights into a range of health conditions.”
The potential benefits extend beyond convenience. By enabling early detection and intervention, voice analysis could reduce healthcare costs and improve outcomes for millions. It also aligns with global efforts to address health disparities, offering a scalable solution for resource-constrained settings.
The implications of this study are profound. Hypertension, a condition often detected too late, could now be screened using a tool as simple as a smartphone app. For individuals in remote or underserved areas, this technology represents a lifeline—a chance to identify risks early and take preventive action.
While further research is needed to validate these findings across larger and more diverse populations, the results are a promising step toward more inclusive healthcare. By harnessing the power of voice, scientists are turning an everyday activity into a key to better health. As the field of vocal biomarkers continues to evolve, its impact on global health could be transformative.
Note: Materials provided above by The Brighter Side of News. Content may be edited for style and length.
Like these kind of feel good stories? Get The Brighter Side of News’ newsletter.
The post AI-powered app accurately detects high blood pressure through voice recordings appeared first on The Brighter Side of News.
Leave a comment
You must be logged in to post a comment.