DocumentCode :
3684754
Title :
Early illness recognition using frequent motif discovery
Author :
Zahra Hajihashemi;Mihail Popescu
Author_Institution :
Dept. of Computer Science, University of Missouri, Columbia, 65203 USA
fYear :
2015
Firstpage :
3699
Lastpage :
3702
Abstract :
Living alone in their own residence, older adults are at risk for late assessment of physical or cognitive changes due to many factors such as their impression that such changes are simply a normal part of aging or their reluctance to admit to a problem. This paper describes an early illness recognition framework using sensor network technology to identify the health trajectory of older adults reflected in patterns of day-today activities. Describing the behavior of older adults could help clinicians to identify those at the greatest risk for functional decline and adverse events. The proposed framework, denoted as Abnormal Frequent Activity Pattern (AFAP), is based on the identification of known past abnormal frequent activities in current sensor data. More specifically, AFAP declares a day abnormal when past frequent abnormal behavior patterns, not found during normal days, are discovered in the current activity data. While AFAP requires the labeling of past days as normal/abnormal, it doesn´t need specific activity identification. Frequent activity patterns (FAP) are found using MEME, a bioinformatics motif detection algorithm. To validate our approach, we used data obtained from TigerPlace, an aging in place community situated in Columbia, MO, where apartments are equipped with sensor networks (motion, bed and depth sensors). A retrospective multiple case study (N=3) design was used to quantify the in-home older adult´s daily routines, over a period of two weeks. Within-person variability of routine activities may be used as a new predictor in the study of health trajectories of older adults.
Keywords :
"Time series analysis","Monitoring","Data mining","Bioinformatics","Databases","Medical services","DNA"
Publisher :
ieee
Conference_Titel :
Engineering in Medicine and Biology Society (EMBC), 2015 37th Annual International Conference of the IEEE
ISSN :
1094-687X
Electronic_ISBN :
1558-4615
Type :
conf
DOI :
10.1109/EMBC.2015.7319196
Filename :
7319196
Link To Document :
بازگشت