TY - JOUR
T1 - To What Extent Can Arm-Hand Skill Performance - of Both Healthy Adults and Children - Be Recorded Reliably Using Multiple Bodily Worn Sensor Devices?
AU - Lemmens, R.J.M.
AU - Seelen, H.A.M.
AU - Timmermans, A.A.A.
AU - Schnackers, M.
AU - Eerden, A.
AU - Smeets, R.J.E.M.
AU - Janssen-Potten, Y.J.M.
PY - 2015/7
Y1 - 2015/7
N2 - Neurological patients often encounter arm-hand problems in daily life. Bodily worn sensors may be used to assess actual performance by quantifying specific movement patterns associated with specific activities. However, signal reliability during activities of daily living should be determined first. The aim is to determine to what extent standardized arm-hand skill performance of both healthy adults and healthy children can be recorded reliably using a combination of multiple sensor devices. Thirty adults (aged > 50 years) and thirty-two children (aged between 6-18 years) performed the activities drinking, eating and combing five times in a standardized setting. Sensor devices, each containing a triaxial accelerometer, gyroscope and magnetometer were attached to the arms, hands and trunk of the participants. Within-subject and between-subject reliability of the signal patterns amongst skill repetitions was determined by calculating Intraclass Correlation Coefficients (ICCs). Median reliability was good to very good for all activities performed (both within and between subjects). Regarding within-subject reliability (instruction-condition), median ICCs ranged between 0.76-0.90 and 0.68-0.92 for the adults and children, respectively. For between-subject reliability (instruction-condition), median ICCs ranged 0.75-0.86 and 0.61-0.90 for the adults and children, respectively. It can be concluded that the above-mentioned sensor system can reliably record activities of daily living in a standardized setting.
AB - Neurological patients often encounter arm-hand problems in daily life. Bodily worn sensors may be used to assess actual performance by quantifying specific movement patterns associated with specific activities. However, signal reliability during activities of daily living should be determined first. The aim is to determine to what extent standardized arm-hand skill performance of both healthy adults and healthy children can be recorded reliably using a combination of multiple sensor devices. Thirty adults (aged > 50 years) and thirty-two children (aged between 6-18 years) performed the activities drinking, eating and combing five times in a standardized setting. Sensor devices, each containing a triaxial accelerometer, gyroscope and magnetometer were attached to the arms, hands and trunk of the participants. Within-subject and between-subject reliability of the signal patterns amongst skill repetitions was determined by calculating Intraclass Correlation Coefficients (ICCs). Median reliability was good to very good for all activities performed (both within and between subjects). Regarding within-subject reliability (instruction-condition), median ICCs ranged between 0.76-0.90 and 0.68-0.92 for the adults and children, respectively. For between-subject reliability (instruction-condition), median ICCs ranged 0.75-0.86 and 0.61-0.90 for the adults and children, respectively. It can be concluded that the above-mentioned sensor system can reliably record activities of daily living in a standardized setting.
KW - Activities of daily living (ADL)
KW - outcome assessment reliability
KW - reproducibility of results
KW - sensor
KW - upper extremity
KW - FOLLOW-UP
KW - RELIABILITY
KW - STROKE
KW - AGREEMENT
KW - QUALITY
U2 - 10.1109/TNSRE.2015.2396082
DO - 10.1109/TNSRE.2015.2396082
M3 - Article
SN - 1534-4320
VL - 23
SP - 581
EP - 590
JO - Ieee Transactions on Neural Systems and Rehabilitation Engineering
JF - Ieee Transactions on Neural Systems and Rehabilitation Engineering
IS - 4
ER -