There is increasing research interest in technologies that can detect grasping, to encourage functional use of the hand as part of daily living, and thus promote upper-extremity motor recovery in individuals with stroke. Force myography (FMG) has been shown to be effective for providing biofeedback to improve fine motor function in structured rehabilitation settings, involving isolated repetitions of a single grasp type, elicited at a predictable time, without upper-extremity movements. The use of FMG, with machine learning techniques, to detect and distinguish between grasping and no grasping, continues to be an active area of research, in healthy individuals. The feasibility of classifying FMG for grasp detection in populations with upper-extremity impairments, in the presence of upper-extremity movements, as would be expected in daily living, has yet to be established. We explore the feasibility of FMG for this application by establishing and comparing (1) FMG-based grasp detection accuracy and (2) the amount of training data necessary for accurate grasp classification, in individuals with stroke and healthy individuals. FMG data were collected using a flexible forearm band, embedded with six force-sensitive resistors (FSRs). Eight participants with stroke, with mild to moderate upper-extremity impairments, and eight healthy participants performed 20 repetitions of three tasks that involved reaching, grasping, and moving an object in different planes of movement. A validation sensor was placed on the object to label data as corresponding to a grasp or no grasp. Grasp detection performance was evaluated using linear and non-linear classifiers. The effect of training set size on classification accuracy was also determined. FMG-based grasp detection demonstrated high accuracy of 92.2% (σ = 3.5%) for participants with stroke and 96.0% (σ = 1.6%) for healthy volunteers using a support vector machine (SVM). The use of a training set that was 50% the size of the testing set resulted in 91.7% (σ = 3.9%) accuracy for participants with stroke and 95.6% (σ = 1.6%) for healthy participants. These promising results indicate that FMG may be feasible for monitoring grasping, in the presence of upper-extremity movements, in individuals with stroke with mild to moderate upper-extremity impairments.
BackgroundIn upper-extremity stroke rehabilitation applications, the potential use of Force Myography (FMG) for detecting grasping is especially relevant, as the presence of grasping may be indicative of functional activity, which is a key goal of rehabilitation. To date, most FMG research has focused on the classification of the raw FMG signal (i.e. instantaneous FMG samples) in order to determine the state of the hand. However, given the temporal nature of force generation during grasping, the use of temporal feature extraction techniques may yield increased accuracy. In this study, the effectiveness of classifying temporal features of the FMG signal for the two-class grasp detection problem of “grasp” versus “no grasp” (i.e. no object in hand) was evaluated with ten healthy participants. The experimental protocol comprised grasp and move tasks, requiring the use of six different grasp types frequently used in daily living, in conjunction with arm and hand movements. Data corresponding to arm and hand movements without grasping were also included to evaluate robustness to false positives. The temporal features evaluated were mean absolute value (MAV), root mean squared (RMS), linear fit (LF), parabolic fit (PF), and autoregressive model (AR). Off-line classification performance of the five temporal features, with a 0.5 s extraction window, were determined and compared to that of the raw FMG signal using area under the receiver operating curve (AUC).ResultsThe raw FMG signal yielded AUC of 0.819 ± 0.098. LF and PF resulted in the greatest increases in classification performance, and provided statistically significant increases in performance. The largest increase obtained was with PF, yielding AUC of 0.869 ± 0.061, corresponding to a 6.1% relative increase over the raw FMG signal. Despite the additional fitting term provided by PF, classification performance did not significantly improve with PF when compared to LF.ConclusionsThe results obtained indicate that temporal feature extraction techniques that derive models of the data within the window may yield modest improvements in FMG based grasp detection performance. In future studies, the use of model-based temporal features should be evaluated with FMG data from individuals with stroke, who might ultimately benefit from this technology.
In this paper an easy-to-use, wearable sensor system capable of deciphering upper-limb based functional tasks associated with stroke rehabilitation is introduced. Such a progress during rehabilitation, and hence can increase the efficiency of the rehabilitation process. The developed system provides quantitative, realactivity. The system is designed to detect the successful completion of functional tasks that involve the grasping, movement, and subsequent release of an object. The developed system consists of an Inertial Measurement Unit (IMU) attached to a band, embedded with force sensitive resistors for extracting Force Myography (FMG) data. The band is wrapped onset of a grasp or release with the use of a Neural Network for data classification. Upon detection of a grasp and subsequent release, the device calculates the distance achieved during the motion using data captured from the attached IMU as a measure of task quality. completion of three functional tasks was evaluated with nine healthy volunteers. The system achieved an average accuracy of 92.67%. Experimental results are presented and discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.