Multimodal wearable sensing for noninvasive diagnosis and severity assessment of aortic stenosis

Aortic stenosis (AS) is a common and progressive valvular heart disease, particularly in older adults, yet diagnosis and longitudinal monitoring remain highly dependent on access to echocardiography and expert interpretation. Scalable, noninvasive tools that can identify AS and assess disease severity outside conventional imaging settings could improve screening, referral, and follow-up pathways.

This presentation will describe the Wearable SENsor to Diagnose and Assess SEverity of Aortic Stenosis (SENSE-AS), a proof-of-concept study recently published in JACC: Advances. The system uses a soft, skin-interfaced wearable platform to acquire synchronized multimodal cardiovascular signals, including electrocardiography, seismocardiography, and phonocardiography. The study evaluated whether sensor-derived features could capture physiologic signatures of AS and correlate with echocardiographic measures of disease severity.

The primary objective was to quantify the relationship between Doppler-derived aortic valve acceleration time and wearable sensor-derived acceleration time. Secondary analyses examined additional multimodal signal features relevant to cardiac timing, mechanical vibration, and acoustic murmur characteristics. By integrating electrical, mechanical, and acoustic cardiovascular information, the platform provides a noninvasive approach for assessing valve-related hemodynamic abnormalities using wearable bioelectronics.

The results support the feasibility of wearable multimodal sensing for identifying and assessing AS severity, with potential applications in outpatient screening, longitudinal monitoring, and precision cardiovascular care. Importantly, this approach may help extend cardiovascular assessment beyond episodic imaging by enabling rapid, low-burden physiological measurements in clinical or remote environments. This work suggests a pathway toward scalable cardiovascular phenotyping, particularly for older adults and populations with limited access to specialty diagnostics. More broadly, the study illustrates how synchronized wearable bioelectronics and interpretable signal analytics may complement established clinical workflows and help advance personalized cardiology.