Machine-learning analysis of Korotkoff sounds distinguished young and older age-defined groups with up to 93.7% accuracy and showed modest but significant association with measured baPWV.
Can machine learning analysis of Korotkoff sounds during routine blood pressure measurement accurately classify arterial stiffness and vascular ageing?
Machine learning analysis of Korotkoff sounds during routine blood pressure measurement can extract features modestly associated with arterial stiffness, offering a potential non-invasive screening tool for vascular ageing.
Effect estimate: Accuracy 93.7% for SqueezeNet in age-defined group classification; Accuracy 89.3% for GoogLeNet in same task; moderate accuracy 87.5% and 81.3% in baPWV-defined task
Background and Objectives Arterial stiffness is a recognized marker of vascular ageing and is associated with adverse cardiovascular outcomes. However, routine assessment of pulse wave velocity (PWV) remains limited in many clinical and home settings. This study investigated the feasibility of extracting arterial stiffness-related information from Korotkoff sounds recorded during cuff-based blood pressure measurement using feature analysis and machine learning. Materials and methods Korotkoff sounds were collected from 123 young (25.9 ± 2.2 years) participants and 112 older (67.5 ± 6.7 years) participants using a custom-developed device as a proof-of-concept for age-related vascular differences. In addition, 81 hospital participants with measured brachial-ankle PWV (baPWV) were enrolled and grouped according to baPWV to further evaluate clinical feasibility. Time- and frequency-domain features were extracted, and both traditional feature-based models and deep learning approaches were applied for classification. Results Extracted features including center of mass, skewness, and peak frequency showed significant differences between the age-stratified groups. Two deep learning models achieved classification accuracies of 89.3% and 93.7%, respectively, outperforming traditional feature-based analysis. In the baPWV-defined classification task, model performance was moderate (accuracy 87.5% and 81.3%). In the baPWV-measured cohort, Korotkoff sound-derived features showed a statistically significant but modest association with measured baPWV. Conclusion Korotkoff sounds contain measurable information related to vascular ageing and arterial stiffness, and machine learning can leverage these signals for group discrimination. Given that the primary comparison used age as a surrogate label and clinical outcomes were not assessed, the present data do not establish incremental value for cardiovascular risk stratification beyond age and blood pressure. Larger studies with standardized PWV measurements, ideally carotid-femoral PWV (cfPWV), and prospective validation are required before prognostic or risk-stratification claims can be made.
Ren et al. (Tue,) conducted a other in vascular ageing and arterial stiffness assessment (n=316). Korotkoff sound analysis with machine learning vs. age-defined groups (young vs older) and baPWV-defined groups (high vs low) was evaluated on classification accuracy of Korotkoff sound analysis distinguishing vascular stiffness-related groups, and correlation with measured baPWV (Accuracy 93.7% for SqueezeNet in age-defined group classification; Accuracy 89.3% for GoogLeNet in same task; moderate accuracy 87.5% and 81.3% in baPWV-defined task). Machine-learning analysis of Korotkoff sounds distinguished young and older age-defined groups with up to 93.7% accuracy and showed modest but significant association with measured baPWV.