Disciplines
Community Health | Health Information Technology | Medical Sciences | Other Mental and Social Health | Psychiatric and Mental Health | Signal Processing
Abstract (300 words maximum)
Aging can be quite an intimidating process for some, especially for those who live alone at home or in an assisted living facility. Although it is a natural progression of life, the stigma around aging as it relates to health is an area of concern for our growing population. Conditions such as dementia and anxiety can often impact older generation which can lead to further deterioration of health. Luckily, we live in an era of digital advancements in both the technology and health sectors that can provide opportunities for research and development such as this – our work focuses on leveraging the strength of vibration sensors and classifiers to detect an activity and categorize based on machine learning. We are using a combination of geophone (seismic sensor), digitizer board, and a single-board computer in an enclosure. All the data collected would be stored locally on the device using influx database and accessed via Grafana. The single board computer is responsible for reading and processing the signals to detect the water vibration and duration of the hygiene events. Once the event has been recorded, we will use machine learning to categorize the hygiene events as activities through feature extraction. The goal behind this research is to automate the process of “detection of a hygiene event” which can assist our healthcare workers to be proactive in detecting a potential health crisis.
Academic department under which the project should be listed
CCSE - Information Technology
Loading...
Primary Investigator (PI) Name
Dr. Maria Valero
Additional Faculty
Dr. Hossain Shahriar, Information Technology, hshahria@kennesaw.edu
Dr. Liang Zhao, Information Technology, lzhao10@kennesaw.edu
Non-Invasive Monitoring of Human Hygiene using Vibration Sensor and Classifiers
Aging can be quite an intimidating process for some, especially for those who live alone at home or in an assisted living facility. Although it is a natural progression of life, the stigma around aging as it relates to health is an area of concern for our growing population. Conditions such as dementia and anxiety can often impact older generation which can lead to further deterioration of health. Luckily, we live in an era of digital advancements in both the technology and health sectors that can provide opportunities for research and development such as this – our work focuses on leveraging the strength of vibration sensors and classifiers to detect an activity and categorize based on machine learning. We are using a combination of geophone (seismic sensor), digitizer board, and a single-board computer in an enclosure. All the data collected would be stored locally on the device using influx database and accessed via Grafana. The single board computer is responsible for reading and processing the signals to detect the water vibration and duration of the hygiene events. Once the event has been recorded, we will use machine learning to categorize the hygiene events as activities through feature extraction. The goal behind this research is to automate the process of “detection of a hygiene event” which can assist our healthcare workers to be proactive in detecting a potential health crisis.