The broader impact/commercial potential of this I-Corps project is to use recent advances in artificial intelligence (AI) and deep learning to enhance public health challenges in nursing homes. Fever, as a non-specific measure of infection, is commonly observed in a broad range of diseases and pandemics. The proposed AI-powered assessment system will create an effective, non-intrusive tool for empowering nursing home facilities and clinics to combat the spread of contagious diseases and future pandemics, all the while providing a higher health resiliency for our communities. The proposed technology creates a real-time health surveillance system that also may be adopted and customized to a wide range of public health applications that require continuous, non-intrusive health monitoring with predictive analytics and proactive decision making. The proposed research has significant opportunities both in the public and private sectors.

This I-Corps project is based on the development of a monitoring system to mitigate the risk and control the spread of epidemic viruses through real-time artificial intelligence, multi-sensor fusion, and video data analytics. In contrast to existing approaches that have a narrow focus with limited intelligence capabilities, the proposed technology offers a holistic solution to enable scalable, reliable symptom assessment and contact tracing from a distance with strict personal privacy measures ensured. By utilizing both red green blue (RGB) and thermal cameras (off-the-shelf products), it may be possible to provide a more precise system that is capable of monitoring several health indicators simultaneously; e.g., body temperature, respiratory rate, coughing, and sneezing while taking a non-intrusive approach. The proposed device is equipped with an AI-enabled contact tracing system for reducing the spread of viruses by identifying the potentially infected individuals at the early stage. For privacy-aware contact tracing, the plan is to leverage previously developed technology for real-time privacy built-in human pose estimation, re-identification, trajectory analysis, and activity recognition. The technology creates lightweight, end-to-end execution of real-time computer vision based on RGB cameras, with the ability to perform at a high frame rate on embedded and edge devices.

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.

Project Start
Project End
Budget Start
2021-02-01
Budget End
2021-07-31
Support Year
Fiscal Year
2021
Total Cost
$50,000
Indirect Cost
Name
University of North Carolina at Charlotte
Department
Type
DUNS #
City
Charlotte
State
NC
Country
United States
Zip Code
28223