University of Delaware

Published purpose

Connected health is one of the most promising applications for collaborative edge computing. Current technology advancement allows for disease or disorders diagnostics, health monitoring and management for older people in a non-clinical at-home and in-community environment by using networked configurable edge devices. However, how to choose the best configurable edge devices to satisfy specific clinical contextunderstanding technologies and support AI-related applications in a dynamic manner are both challenges. The proposed project aims to provide transformative solutions to solve these two challenges and assist in the execution of proposed deep learning algorithms for remote contactless sensing on single and multiple edge devices for connected health applications. The proposed project will (1) identify suitable edge devices and propose edge architecture for NRC’s contactless sensing technology and software development needs, and (2) design, develop and support efficient and optimized porting of NRC’s technology to the proposed edge infrastructure. At the end of the project, the team will deliver a network of edge devices for the passive smart sensing system for a remote at-home use scenario.