Event sponsored by:
AI Health
+DataScience (+DS)
Biomedical Engineering (BME)
Biostatistics and Bioinformatics
Center for Computational Thinking
Computer Science
CTSI CREDO
Duke Clinical and Translational Science Award (CTSA)
Duke Clinical and Translational Science Institute (CTSI)
Electrical and Computer Engineering (ECE)
Mechanical Engineering and Materials Science (MEMS)
Contact:
Duke AI HealthSpeaker:
Xiaoyue Ni, PhD
Human interactions with the environment and with others generate mechano-acoustic (MA) signals that encode rich information about physiological and biomechanical processes. Although these signals often attenuate at the skin-air interface, epidermal electronics provide a powerful means to capture them, offering a transformative approach for non-invasive, real-time health monitoring. We present a soft, skin-mounted MA sensing platform that integrates inertial measurement units, strain gauges, or ultrasound transducers to record body sounds and kinematics with high fidelity. This thin, conformal "sticker" enables MA detection at anatomically challenging sites inaccessible to rigid, planar wearables. Paired with frequency-domain analysis and machine-learning algorithms, the system interprets high-density, networked MA data streams to infer a wide spectrum of health and performance metrics. Demonstrated applications include wireless, continuous ambulatory monitoring of vital signs (e.g., heart rate, respiration, activity), tissue stiffness, muscle force, body morphology, posture, and unconventional biomarkers such as speech patterns, cough frequency, swallowing, social interaction metrics, and sleep quality. Ongoing work extends these capabilities toward quantifying proprioception through biomechanical signature analysis and advancing flexible, stretchable ultrasound systems for brain imaging and stimulation. Ultimately, this technology aims to capture a high-dimensional array of mechanical and acoustic signatures, enabling comfortable, accurate, and comprehensive decoding of physiological state, behavioral patterns, functional performance, and cognitive or intentional states in real time.
AI Health Virtual Seminar Series