A new scheme for the development of IMU-based activity recognition systems for telerehabilitation.

Wearable human activity recognition systems (HAR) using inertial measurement units (IMU) play a key role in the development of smart rehabilitation systems. Training of a HAR system with patient data is costly, time-consuming, and difficult for the patients. This study proposes a new scheme for the optimal design of HARs with minimal involvement of the patients. It uses healthy subject data for optimal design for a set of activities used in the rehabilitation of PD1 patients. It maintains its performance for individual PD subjects using a single session data collection and an adaptation procedure. In the optimal design, several classifiers (i.e. NM, k-NN, MLP with RBF as a hidden layer, and multistage RBF SVM) were investigated. Features were signal-based in the time, frequency, and time-frequency domains. Double-stage feature extraction by PCA and fisher technique was used. The optimal design reached a recall of 95% on healthy subjects using only two sensors on the left thigh and forearm. Implementing the adaptation procedure on two PD subjects, the performance was maintained above 80%. Post analysis on the performance of the adapted HAR showed a slight drop in precision (above 87% to above 81%) for activities that was performed in sitting condition.

Increased REM sleep without atonia in Parkinson disease with freezing of gait.

OBJECTIVE: The objective of this cross-sectional study was to test the hypothesis that patients with Parkinson disease (PD) and freezing of gait (PD+FOG) would demonstrate sleep disturbances comparable to those seen in patients with REM sleep behavior disorder (RBD) and these changes would be significantly different from those in PD patients without FOG (PD-FOG) and age-matched controls. METHODS: We conducted overnight polysomnography studies in 4 groups of subjects: RBD, PD-FOG, PD+FOG, and controls. Tonic and phasic muscle activity during REM sleep were quantified using EMG recordings from the chin, compared among study groups, and correlated with disease metrics. RESULTS: There were no significant differences in measures of disease severity, duration, or dopaminergic medications between the PD+FOG and PD-FOG groups. Tonic muscle activity was increased significantly (p < 0.007) in the RBD and PD+FOG groups compared to the PD-FOG and control groups. There was no significant difference in tonic EMG between the PD+FOG and RBD group (p = 0.364), or in tonic or phasic EMG between the PD-FOG and control group (p = 0.107). Phasic muscle activity was significantly increased in the RBD group compared to all other groups (p = 0.029) and between the PD+FOG and control group (p = 0.001), but not between the PD+FOG and PD-FOG groups (p = 0.059). CONCLUSIONS: These findings provide evidence that increased muscle activity during REM sleep is a comorbid feature of patients with PD who exhibit FOG as a motor manifestation of their disease.

Comparison of neuromuscular abnormalities between upper and lower extremities in hemiparetic stroke.

We studied the neuromuscular mechanical properties of the elbow and ankle joints in chronic, hemiparetic stroke patients and healthy subjects. System identification techniques were used to characterize the mechanical abnormalities of these joints and to identify the contribution of intrinsic and reflex stiffness to these abnormalities. Modulation of intrinsic and reflex stiffness with the joint angle was studied by applying PRBS perturbations to the joint at different joint angles. The experiments were performed for both spastic (stroke) and contralateral (control) sides of stroke patients and one side of healthy (normal) subjects. We found reflex stiffness gain (GR) was significantly larger in the stroke than the control side for both elbow and ankle joints. GR was also strongly position dependent in both joints. However, the modulation of GR with position was slightly different in two joints. GR was also larger in the control than the normal joints but the differences were significant only for the ankle joint. Intrinsic stiffness gain (K) was also significantly larger in the stroke than the control joint at elbow extended positions and at ankle dorsiflexed positions. Modulation of K with the ankle angle was similar for stroke, control and normal groups. In contrast, the position dependency of the elbow was different. K was larger in the control than normal ankle whereas it was lower in the control than normal elbow. However, the differences were not significant for any joint. The findings demonstrate that both reflex and intrinsic stiffness gain increase abnormally in both upper and lower extremities. However, the major contribution of intrinsic and reflex stiffness to the abnormalities is at the end of ROM and at the middle ROM, respectively. The results also demonstrate that the neuromuscular properties of the contralateral limb are not normal suggesting that it may not be used as a suitable control at least for the ankle study.

Mechanical abnormalities of the spastic ankle in chronic stroke subjects.

We examined the intrinsic and reflex contributions to ankle stiffness in people with chronic stroke and healthy subjects using the parallel system identification technique. Modulation of intrinsic and reflex stiffness was characterized by applying pseudorandom binary sequence (PRBS) perturbations to the ankle at different initial ankle joint over the entire range of motion (ROM). The experiments were performed for both paretic (stroke) and contralateral (control) side. Healthy (normal) subjects were used a secondary control. Reflex stiffness gain significantly increased in stroke than in control side at most positions. Intrinsic stiffness gain also increased significantly at dorsiflexing positions. These changes were position dependent. Thus, the abnormalities in intrinsic stiffness gain increased continuously from middle plantarflexion to full dorsiflexion while the major increase in reflex stiffness happened at the middle ROM. No significant changes were found in other intrinsic and reflex stiffness parameters. As compared to the normal ankle, the reflex stiffness gain of the control side was significantly larger, indicating that the control side is not normal. These findings demonstrate that both intrinsic and reflex stiffness contribute significantly to the mechanical abnormality associated with spastic ankle in hemiplegic stroke subjects. The results also suggest that the contralateral limb may not be used as a suitable control.