I-Corps: Translation Potential of a Novel Ultra-low-power Wearable Health Monitoring System

Award ID: 2504722

This I-Corps project is based on the lab to market translation of a novel wearable health monitoring device that provides continuous, real-time monitoring of vital signs and early detection of health-critical events. This technology provides a cost-effective and energy-efficient solution that enhances accessibility to real-time health data for individuals with anxiety disorders, cardiovascular conditions, and other medical concerns requiring continuous monitoring. This solution addresses the lack of affordable, real-time, and continuous health monitoring for individuals at risk of critical health events, such as panic attacks or cardiac irregularities. Current wearable solutions either lack accuracy due to motion interference or require high power consumption, limiting their usability for long-term monitoring. With the U.S. facing an increasing prevalence of chronic health conditions and a growing demand for remote patient monitoring, there is need for an energy-efficient, reliable, and scalable solution that bridges the gap between clinical-grade monitoring and consumer accessibility. Commercializing this solution has the potential to benefit society and the economy by reducing healthcare costs, improving early intervention outcomes, and expanding access to personalized health tracking, which can ultimately enhance public health and well-being. This I-Corps project utilizes experiential learning and first-hand industry engagement to assess the commercialization potential of an ultra-low-power wearable health monitoring system that integrates advanced biometric sensors, near-sensor artificial intelligence (AI) processing, and novel signal processing techniques. This solution consists of a hybrid computational framework that leverages emerging stochastic computing (SC) and hyperdimensional computing (HDC) to reduce energy consumption while enhancing real-time signal processing. Unlike conventional wearables, which rely on high-power computation, this approach enables efficient near-sensor AI processing, minimizing data transmission delays and computational overhead. Additionally, the fusion of multi-modal sensor data with deep learning models improves signal integrity and detection accuracy for health-critical events. The benefits of this approach include longer battery life, lower manufacturing costs, and enhanced accuracy in detecting physiological anomalies. By eliminating energy-intensive digital conversions and leveraging low-complexity AI models, this technology ensures scalability and affordability for widespread adoption in clinical, consumer, and remote patient monitoring applications. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Ruth Shuman