Improving independent mobility in post-stroke patients is one of the main goals of most rehabilitation strategies. While quantitative gait assessment is crucial to provide a meaningful feedback on the recovery progress, the irregularity of hemiparetic walking prevents the use of classical Inertial Measurement Unit (IMU)-based gait analysis algorithms. In this paper, we propose a novel low-cost system, which relies on a single wearable IMU attached to the lower trunk, to estimate spatio-temporal gait parameters of both hemiparetic and healthy subjects. A new procedure for temporal features' computation and two modified versions of well-known step length (i.e., spatial features) estimators are derived. In both cases, we exploit dynamic calibration constants, related to the power of an individual gait pattern, to deal with the typical asymmetry and inter-subject variability of hemiparetic gait. The spatio-temporal features estimated with the proposed methods are compared with ground-truth parameters extracted by an optoelectronic system. The obtained results show very high correlations between estimated and reference values.
Accurate gait analysis in post-stroke patients using a single inertial measurement unit / Parisi, Federico; Ferrari, Gianluigi; Baricich, Alessio; D'Innocenzo, Marco; Cisari, Carlo; Mauro, Alessandro. - (2016), pp. 335-340. (Intervento presentato al convegno 13th Annual Body Sensor Networks Conference, BSN 2016 tenutosi a University of California, San Francisco (UCSF) Mission Bay Conference Center, usa nel 2016) [10.1109/BSN.2016.7516284].
Accurate gait analysis in post-stroke patients using a single inertial measurement unit
PARISI, Federico;FERRARI, Gianluigi;
2016-01-01
Abstract
Improving independent mobility in post-stroke patients is one of the main goals of most rehabilitation strategies. While quantitative gait assessment is crucial to provide a meaningful feedback on the recovery progress, the irregularity of hemiparetic walking prevents the use of classical Inertial Measurement Unit (IMU)-based gait analysis algorithms. In this paper, we propose a novel low-cost system, which relies on a single wearable IMU attached to the lower trunk, to estimate spatio-temporal gait parameters of both hemiparetic and healthy subjects. A new procedure for temporal features' computation and two modified versions of well-known step length (i.e., spatial features) estimators are derived. In both cases, we exploit dynamic calibration constants, related to the power of an individual gait pattern, to deal with the typical asymmetry and inter-subject variability of hemiparetic gait. The spatio-temporal features estimated with the proposed methods are compared with ground-truth parameters extracted by an optoelectronic system. The obtained results show very high correlations between estimated and reference values.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.