TY - JOUR
AU - Poorani, Dhivya
AB - AbstractThe increase in demand of remote health monitoring for elderly motivates the development of a novel framework for decision making under realistic conditions. Existing works on health monitoring focuses more on sensor development, security, data collection and feature extraction. But, there is no solution for analyzing multidimensional physiological parameters and activity recognition integrated in a single system. Existing systems compromises either on accuracy or detection time. To overcome these issues, this paper proposes a new idea for an automated Sensor based Decision Making (SbDM) framework that categorizes activity, vital parameters and uses Open Geo-spatial Consortium (OGC) standard for web enablement of sensor data. Two schemes such as abnormality detection in vital parameters and activity recognition are proposed in SbDM. SbDM combine generative Hidden Markov Model for activity prediction and discriminative Adaptive Neuro Fuzzy Inference System for accurate activity classification. These two modules are combined for early detection and accurate activity recognition. Combined activity recognition systems help to improve the reliability of the system and reduce false decision. SbDM also implements Decision Tree Pruning for abnormality detection in vital parameters. Activity and vital parameters measurements are closely related for future decision making or alert generation.
TI - Sensor based efficient decision making framework for remote healthcare
JF - Journal of Ambient Intelligence and Smart Environments
DO - 10.3233/ais-150330
DA - 2015-07-13
UR - https://www.deepdyve.com/lp/ios-press/sensor-based-efficient-decision-making-framework-for-remote-healthcare-QMjtZf0xdl
SP - 461
EP - 481
VL - 7
IS - 4
DP - DeepDyve
ER -