II-NEW: Infrastructure for Research and Education in Machine Perception and Sensory Motor Learning for Home Health Care

Delaware State University, Dover DE

Investigators

Abstract

This project provides infrastructure to support research concerning the investigation, design, and integration of reusable sensory-motor services that proactively monitor patient health status and provide flexible tools for intervention. The shortage of skilled allied healthcare professionals and their finite capacity prevents the continuous monitoring of patient vital statistics. Leveraging an aging population?s preference to remain at home while managing chronic illness, this project enables research into the use of machine perception and sensory motor learning for recognition and intervention of degrading health status in a home care environment. The investigators are building a new infrastructure consisting of compute nodes, robotic platforms, sensing systems, and novel service framework for research and education in Machine Perception and Sensory Motor Learning for healthcare applications at Delaware State University. By factoring perceptual systems and making them accessible services on the network, a robot becomes a virtual assemblage that recruits arbitrary resources at run-time. Such service compositions provide a richer suite of perceptual and robotic services because it draws from the power-set of available sensory-motor apparatus. Service compositions include methods, such as ensembles in machine learning, that construct high fidelity models for pattern learning by combining a number of simpler component models. The infrastructure built from this project enables research in many areas, such as perceptual computing, robotics, and machine learning. The broader impact of the project includes more proactive tools for detection and intervention in healthcare and training of undergraduate and graduate students (including underrepresented students in STEM) in interdisciplinary research.