Future Technology For E Health Systems

1372 Words6 Pages
Abstract Lot of research has been done on Human Activity Recognition(HAR) in the last decade and HAR is expected to be future technology for e-health systems. Here, I discuss various models built using SAS® Enterprise Miner™ 14.1, on a free public domain dataset containing165,633 observations and 19 attributes and compare each model with another. Data used in this discussion, represents the metrics of accelerometers mounted on waist, left thigh, right arm and right ankle of 4 individuals performing five different activities recorded over a period of eight hours. Finally, I propose a Stepwise Logistic-fed-AutoNeural model to recognize human activity. The accuracy of the model is 98.73% Introduction The advances in medical care and rise in…show more content…
There are prospects of developing assistive technologies to support care of the older adults using research work on HAR. E-Health systems like AAL can be developed by using patients ' routine data provided by activity recognition. The most common methods used to recognize human activity are image processing and usage of wearable sensors. Though image processing doesn 't require patient to wear any equipment, it has some major problems such as requiring installation of cameras and good light. In addition, its operations are restricted to indoor environments and users have privacy concerns. The use of wearable sensors has addressed all these problems, but requires wearing of equipment by the user for long durations[1]. In this paper, I discuss various models built using SAS® Enterprise Miner™ 14.1, on a free public domain dataset containing 165,633 observations and 19 attributes and compare each model with another . Data used in this discussion, represents the metrics of accelerometers mounted on waist, left thigh, right arm, and right ankle of 4 subjects performing five different activities. These activities, such as sitting-down, standing-up, standing, walking, and sitting, recorded over a period of eight hours of four healthy subjects is used for our analysis. Data Data was collected from accelerometers mounted on waist, left thigh, right arm, and right ankle of 4 subjects while they were performing five different activities. This data was
Open Document