Authors:
Ana Cristina Marcén
1
;
Jesús Carro
2
and
Violeta Monasterio
1
Affiliations:
1
Universidad San Jorge, Spain
;
2
Universidad San Jorge, University of Zaragoza, CIBER in Bioengineering and Biomaterials and Nanomedicine (CIBER-BBN), Spain
Keyword(s):
Wearable Computers, Pervasive Health, Support Vector Machines, Nocturnal Agitation, Accelerometry.
Related
Ontology
Subjects/Areas/Topics:
Mobile and Pervasive Computing
;
Mobile Software and Services
;
Pervasive Health
;
Telecommunications
;
Wearable Computers
;
Wireless Information Networks and Systems
Abstract:
Nocturnal agitation is one of the symptoms exhibited by dementia patients. Diagnosing and monitoring the
evolution of agitation is difficult because patient monitoring requires a doctor, nurse or caregiver observing
patients for extended periods of time. In this work, we propose to use an automatic monitoring system based
on wearable technology that complements the caregiver’s work. The proposed system uses a wrist wearable
device to record agitation data, which are subsequently classified through machine learning techniques as
quantifiable indexes of nocturnal agitation. Preliminary tests performed with volunteers showed that the classification
of recorded movements between nocturnal agitation or quiet periods was successful in 78.86% of the
cases. This proof of concept demonstrates the feasibility of using wearable technology to monitor nocturnal
agitation.