Immersive Virtual Reality During Robot-assisted Walking.

STUDY DESIGN: an experimental, 2-group, single-center trial will be conducted in which healthy participants will perform one VR-enhanced RAG session in which 3 different OF speeds will be presented in a random order: 1) walking with slow OF speed, 2) walking with matched OF speed, and 3) walking with fast OF speed. The VE will be different for both groups. The experimental group will have a stimulus rich VE that represents a park where subjects will walk through. This VE will be more interactive by allowing other avatars or animals to walk through the VE. The control group will have a stimulus poor VE that represents an endless hallway. Since the subjects will walk alone in the hallway for the entire time, this VE will be less interactive.

PATIENT RECRUITMENT: we aim to recruit 28 healthy participants, divided equally into two groups. In case subjects drop out, additional participants will be tested till 14 participants in each group completed all conditions.

PROCEDURE: Participants will be tested during one VR-enhanced RAG session at the rehabilitation center TrainM (Antwerp). Prior to the start of the experiment, surface electrodes will be placed bilateral on the Mm. rectus femoris and Mm. biceps femoris to measure muscle activity. Electrode placement will follow as closely as possible the guidelines of SENIAM. The skin underlying the electrode will be shaved and cleaned to improve electrode-skin contact and reduce impedance. At the start of the session, participants will be habituated to walking in the Lokomat without the VR. During this habituation trial, the appropriate settings to walk comfortable in the exoskeleton will be determined. To make sure all participants will walk comfortable in the Lokomat, subjects will be provided with a body weight support (BWS) of 30%. Guidance Force (GF) will be set on 80% for all participants. With lower levels of GF, the biofeedback values will show smaller changes and can become negligible. Lastly, the walking speed (WS) will be set at 2.8 km/h for all participants. Preparing and installing the participant and determining the appropriate Lokomat settings will take no longer than 20 minutes. When the Lokomat settings are defined, the Lokomat will be stopped for a moment so that the VR can be added. Participants will put on the head-mounted display (HMD) and will start walking again for one minute while watching the VE scene expanding at a speed matching their walking speed to get used to the VR. After the habituation trial, participants will walk in the Lokomat with the three different OF speeds that will be presented in a random order for 7 minutes each:

1. slower OF: 0.5 times the individual's WS,
2. matched OF: same as the individual's WS,
3. faster OF: 2 times the individual's WS.

In order to normalize the registered EMG signals, the maximum voluntary isometric contraction (MVIC) test will be used. Each subject carries out the MVIC test 3 times for each muscle. Subjects must reach their maximum strength as quickly as possible and must keep it for 6 seconds. Subjects will perform the MVIC test in the following testing positions: prone with the knees in 30° of flexion (Mm. biceps femoris); seated with the knees in 60°of flexion, and the hips in 90° of flexion (Mm. rectus femoris). The average of the three tests will be taken for each muscle.

RANDOMIZATION: Subjects who agree to participate in this study will be equally randomized in a 1:1 ratio into 2 groups. For randomization, sealed envelopes will be prepared in advance and marked on the inside with a VE + or VE -. During the intervention, the speed of the OF (slow, matched or fast) will also be randomized. To participants, it will only be said that they will have to walk in the Lokomat with the VR for 21 minutes. They will not know that and when the speed of the OF will change.

MATERIALS: The VR will be provided with the use of a novel HMD VR system 'Oculus Rift' (Oculus Rift, LLC, US). The Oculus Rift promises the user more immersion and visual clarity of the virtual rendered world, by use of a custom display and optics technology designed specifically for VR.

Surface EMG electrodes will be used to measure the muscle activity of the Mm. rectus femoris and Mm. biceps femoris. Surface EMG data will be captured with the ME6000 EMG-system (Mega Electronics Ltd) using Conmed Cleartrace electrodes (Ag/AgCl) with an active diameter of 10mm.

The Lokomat has a biofeedback system that measures and displays participants' activity in real-time. With the use of force sensors, this biofeedback system measures the torques acting onto the corresponding Lokomat joint. This is calculated from the signals of the force sensors in the drives. Positive values correspond to external forces (from the subject) acting into flexion direction, while negative values correspond to external forces (from the subject) acting into extension. Besides force sensors, the Lokomat also comprises position sensors. These sensors will be used to measure the range of motion (ROM) of the hip- and knee joints. A value of 0 corresponds to a neutral position, positive values corresponds to flexion and negative values to extension.

STATISTICAL ANALYSIS: Statistics will be performed using SPSS (IBM, Chicago, IL). The significance level will be set on 5%. Descriptive statistics will be calculated for baseline participant characteristics. Means and standard deviations will be calculated for continues variables and frequencies and percentages for categorical variables. Active participation will be determined with the use of muscle activity (EMG), interaction torques and range of motion and will be measured continuously during the session. Data analysis of the average of every minute will be conducted. To investigate the effect of time, the different OF speeds, and the different VEs, repeated-measures ANOVA (within-between subject design) will be analyzed.