Real-time motion onset recognition for robot-assisted gait rehabilitation.

BACKGROUND: Many patients with neurological movement disorders fear to fall while performing postural transitions without assistance, which prevents them from participating in daily life. To overcome this limitation, multi-directional Body Weight Support (BWS) systems have been developed allowing them to perform training in a safe environment. In addition to overground walking, these innovative/novel systems can assist patients to train many more gait-related tasks needed for daily life under very realistic conditions. The necessary assistance during the users' movements can be provided via task-dependent support designs. One remaining challenge is the manual switching between task-dependent supports. It is error-prone, cumbersome, distracts therapists and patients, and interrupts the training workflow. Hence, we propose a real-time motion onset recognition model that performs automatic support switching between standing-up and sitting-down transitions and other gait-related tasks (8 classes in total). METHODS: To predict the onsets of the gait-related tasks, three Inertial Measurement Units (IMUs) were attached to the sternum and middle of outer thighs of 19 controls without neurological movement disorders and two individuals with incomplete Spinal Cord Injury (iSCI). The data of IMUs obtained from different gait tasks was sent synchronously to a real-time data acquisition system through a custom-made Bluetooth-EtherCAT gateway. In the first step, data was applied offline for training five different classifiers. The best classifier was chosen based on F1-score results of a Leave-One-Participant-Out Cross-Validation (LOPOCV), which is an unbiased way of testing. In a final step, the chosen classifier was tested in real time with an additional control participant to demonstrate feasibility for real-time classification. RESULTS: Testing five different classifiers, the best performance was obtained in a single-layer neural network with 25 neurons. The F1-score of [Formula: see text] and [Formula: see text] are achieved on testing using LOPOCV and test data ([Formula: see text], participants = 20), respectively. Furthermore, the results from the implemented real-time classifier were compared with the offline classifier and revealed nearly identical performance (difference = [Formula: see text]). CONCLUSIONS: A neural network classifier was trained for identifying the onset of gait-related tasks in real time. Test data showed convincing performance for offline and real-time classification. This demonstrates the feasibility and potential for implementing real-time onset recognition in rehabilitation devices in future.

Automated stand-up and sit-down detection for robot-assisted body-weight support training with the FLOAT.

Patients with impaired walking function are often dependent on assistive devices to retrain gait and regain independence in life. To provide adequate support, gait rehabilitation devices have to be manually set to the correct support mode or have to recognize the type and starting point of a certain motion automatically. For automated motion type detection, machine learning-based classification algorithms using sensor signals from different body parts can achieve robust performance. However, until today, there is only little work available to detect motion onset. In this paper, we investigate task onset detection of sit-to-stand and stand-to-sit transitions. The focus of the current study is twofold: First, the optimal window size for the online classification algorithm shall be found. Second, the ideal sensor placement in a single sensor-setup, to detect movement onset with shortest detection delays possible is of interest. For our investigations a linear discriminant analysis classifier, basic kinematic features, and a leave-one-subject-out cross validation are used. As a result, an average detection time of 56 milliseconds (SD 111) for sit-to-stand and 48 milliseconds (SD 137) for stand-to-sit were achieved with a window size of 15 and 35 milliseconds respectively at a data rate of 200 hertz. For sit-to-stand transitions, a sensor close to the tenth vertebra and for stand-to-sit transitions close to the posterior pelvis provided the smallest detection times.

The Effect of Haptic Guidance on Learning a Hybrid Rhythmic-Discrete Motor Task.

Bouncing a ball with a racket is a hybrid rhythmic-discrete motor task, combining continuous rhythmic racket movements with discrete impact events. Rhythmicity is exceptionally important in motor learning, because it underlies fundamental movements such as walking. Studies suggested that rhythmic and discrete movements are governed by different control mechanisms at different levels of the Central Nervous System. The aim of this study is to evaluate the effect of fixed/fading haptic guidance on learning to bounce a ball to a desired apex in virtual reality with varying gravity. Changing gravity changes dominance of rhythmic versus discrete control: The higher the value of gravity, the more rhythmic the task; lower values reduce the bouncing frequency and increase dwell times, eventually leading to a repetitive discrete task that requires initiation and termination, resembling target-oriented reaching. Although motor learning in the ball-bouncing task with varying gravity has been studied, the effect of haptic guidance on learning such a hybrid rhythmic-discrete motor task has not been addressed. We performed an experiment with thirty healthy subjects and found that the most effective training condition depended on the degree of rhythmicity: Haptic guidance seems to hamper learning of continuous rhythmic tasks, but it seems to promote learning for repetitive tasks that resemble discrete movements.