TY - GEN
T1 - A system for assessment of limb movements in sleep
AU - Adami, Adriana M.
AU - Adami, Andre G.
AU - Hayes, Tamara L.
AU - Beattie, Zachary T.
N1 - Copyright:
Copyright 2014 Elsevier B.V., All rights reserved.
PY - 2013
Y1 - 2013
N2 - Movement in bed may be an indicator of health problems, i.e., the alteration of the pattern or amount of motor activity can be a disease marker. It can reflect illnesses ranging from flu to depression, pain, or the side effects of certain treatments. There are also motor disturbances that are triggered by sleep such as restless legs syndrome and periodic limb movements during sleep that reduce sleep quality. The assessment of nocturnal motor disturbances in sleep is traditionally performed through overnight polysomnography or actigraphy. In this paper, we investigate the use of unobtrusive sensors (load cells) installed under the supports of a bed to assess movements in bed. Body and leg movements are discriminated using a linear classifier with features derived from load cell signals. Leg movements are scored as periodic leg movements using the criteria defined by The World Association of Sleep Medicine, and a periodic leg movement index is estimated. The system is validated against technicians' annotations of body movements and leg movements (scorings from EMG data) collected from 17 patients during a one night polysomnogram exam. The classification rate is equal to 97%, and the Spearman's correlation coefficient between the periodic leg movement indexes estimated by the system and those obtained from polysomnogram is 0.927. These results demonstrate the feasibility of using load cells for movement classification and detection of periodic leg movements.
AB - Movement in bed may be an indicator of health problems, i.e., the alteration of the pattern or amount of motor activity can be a disease marker. It can reflect illnesses ranging from flu to depression, pain, or the side effects of certain treatments. There are also motor disturbances that are triggered by sleep such as restless legs syndrome and periodic limb movements during sleep that reduce sleep quality. The assessment of nocturnal motor disturbances in sleep is traditionally performed through overnight polysomnography or actigraphy. In this paper, we investigate the use of unobtrusive sensors (load cells) installed under the supports of a bed to assess movements in bed. Body and leg movements are discriminated using a linear classifier with features derived from load cell signals. Leg movements are scored as periodic leg movements using the criteria defined by The World Association of Sleep Medicine, and a periodic leg movement index is estimated. The system is validated against technicians' annotations of body movements and leg movements (scorings from EMG data) collected from 17 patients during a one night polysomnogram exam. The classification rate is equal to 97%, and the Spearman's correlation coefficient between the periodic leg movement indexes estimated by the system and those obtained from polysomnogram is 0.927. These results demonstrate the feasibility of using load cells for movement classification and detection of periodic leg movements.
KW - load cells
KW - movement classification
KW - movement in sleep
KW - periodic leg movements
UR - http://www.scopus.com/inward/record.url?scp=84894140051&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84894140051&partnerID=8YFLogxK
U2 - 10.1109/HealthCom.2013.6720712
DO - 10.1109/HealthCom.2013.6720712
M3 - Conference contribution
AN - SCOPUS:84894140051
SN - 9781467358019
T3 - 2013 IEEE 15th International Conference on e-Health Networking, Applications and Services, Healthcom 2013
SP - 419
EP - 423
BT - 2013 IEEE 15th International Conference on e-Health Networking, Applications and Services, Healthcom 2013
T2 - 2013 IEEE 15th International Conference on e-Health Networking, Applications and Services, Healthcom 2013
Y2 - 9 October 2013 through 12 October 2013
ER -