TY - JOUR
T1 - Affordable, automatic quantitative fall risk assessment based on clinical balance scales and Kinect data
AU - Colagiorgio, P.
AU - Romano, F.
AU - Sardi, F.
AU - Moraschini, M.
AU - Sozzi, A.
AU - Bejor, M.
AU - Ricevuti, G.
AU - Buizza, A.
AU - Ramat, S.
PY - 2014
Y1 - 2014
N2 - The problem of a correct fall risk assessment is becoming more and more critical with the ageing of the population. In spite of the available approaches allowing a quantitative analysis of the human movement control system's performance, the clinical assessment and diagnostic approach to fall risk assessment still relies mostly on non-quantitative exams, such as clinical scales. This work documents our current effort to develop a novel method to assess balance control abilities through a system implementing an automatic evaluation of exercises drawn from balance assessment scales. Our aim is to overcome the classical limits characterizing these scales i.e. limited granularity and inter-/intra-examiner reliability, to obtain objective scores and more detailed information allowing to predict fall risk. We used Microsoft Kinect to record subjects' movements while performing challenging exercises drawn from clinical balance scales. We then computed a set of parameters quantifying the execution of the exercises and fed them to a supervised classifier to perform a classification based on the clinical score. We obtained a good accuracy (~82%) and especially a high sensitivity (~83%).
AB - The problem of a correct fall risk assessment is becoming more and more critical with the ageing of the population. In spite of the available approaches allowing a quantitative analysis of the human movement control system's performance, the clinical assessment and diagnostic approach to fall risk assessment still relies mostly on non-quantitative exams, such as clinical scales. This work documents our current effort to develop a novel method to assess balance control abilities through a system implementing an automatic evaluation of exercises drawn from balance assessment scales. Our aim is to overcome the classical limits characterizing these scales i.e. limited granularity and inter-/intra-examiner reliability, to obtain objective scores and more detailed information allowing to predict fall risk. We used Microsoft Kinect to record subjects' movements while performing challenging exercises drawn from clinical balance scales. We then computed a set of parameters quantifying the execution of the exercises and fed them to a supervised classifier to perform a classification based on the clinical score. We obtained a good accuracy (~82%) and especially a high sensitivity (~83%).
UR - http://www.scopus.com/inward/record.url?scp=84945306065&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84945306065&partnerID=8YFLogxK
U2 - 10.1109/EMBC.2014.6944377
DO - 10.1109/EMBC.2014.6944377
M3 - Article
C2 - 25570745
VL - 2014
SP - 3500
EP - 3503
JO - Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference
JF - Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference
SN - 1557-170X
ER -