Objective: To investigate the clinical impact of brain-machine interface (BMI) training integrated with physical rehabilitation in paraplegic patients with spinal cord injury, specifically examining upright posture support using a stand-in-table device with EEG-based control of a virtual avatar and visuo-tactile feedback
Gather these items before starting the experiment. Check off items as you prepare.
Autodesk • N/A
As an Amazon Associate, we earn from qualifying purchases. Product links help support this free resource.
Eight paraplegic patients with chronic spinal cord injury (>1 year) were enrolled in the Walk Again Neurorehabilitation (WA-NR) protocol. All participants signed written informed consent before enrolling in the study. Protocol was approved by local ethics committee and Brazilian federal government ethics committee.
Note: Seven patients had complete SCI, one had incomplete SCI. All research activities carried out in accordance with guidelines and regulations of Associação de Assistência à Criança Deficiente and CONEP.
“Eight paraplegic patients, suffering from chronic (>1 year) spinal cord injury (SCI, seven complete and one incomplete)”
Comprehensive clinical assessments performed on Day 0 (first day of training) including: ASIA Impairment Scale, Semmes-Weinstein Monofilament Test, temperature/vibration/proprioception/deep pressure sensitivity evaluation, muscle strength test (Lokomat L-force Evaluation), Thoracic-Lumbar Scale for trunk control, WISCI, SCIM, McGill Pain Questionnaire, Visual Analogue Scale for pain, range of motion assessment, Modified Ashworth Scale, Lokomat L-stiff Evaluation for spasticity, WHOQoL-Bref, Rosenberg Self-Esteem Scale, and Beck Depression Inventory.
Note: Routine general clinical evaluations also performed before and after every activity including cardiovascular function, intestinal and urinary emptying, skin inspection, and spasticity handling
“Such clinical evaluation started on the first day patients began training (Day 0), and were repeated after 4, 7, 10, and 12 months”
Patient sits in immersive virtual reality environment wearing Oculus Rift headset. Patient imagines movement of arms to modulate EEG activity (recorded via 16-channel EEG) to generate high-level motor commands such as 'walk' or 'stop'. Virtual avatar movements are displayed from first-person perspective. Patient receives visuo-tactile feedback via haptic display (vibrator arrays) on forearms synchronized with virtual foot rolling on ground. After selecting correct state, patient confirms choice by performing isometric contraction of triceps muscle.
Note: This is the initial BMI strategy. Patients progress to leg imagination once they master this method.
“an immersive virtual reality environment in which a seated patient employed his/her brain activity, recorded via a 16-channel EEG, to control the movements of a human body avatar, while receiving visuo-tactile feedback”
Once patients master arm imagination method, they learn to use EEG signals to control individual avatar leg stepping by imagining movements of their own legs. Virtual avatar is simulated in MotionBuilder 2014 and visualized from first-person perspective using Oculus Rift. Patients continue to receive visuo-tactile feedback via haptic display on forearms synchronized with virtual foot rolling.
Note: This represents progression to more specific motor imagery control
“they learned to use EEG signals to control individual avatar/robotic leg stepping by imagining movements of their own legs”
Patient stands upright supported by stand-in-table device. Patient uses identical BMI protocol and virtual environment as seated condition, controlling virtual avatar movements via EEG-based brain activity (16-channel EEG recording). Patient receives visuo-tactile feedback via haptic display on forearms. Patient imagines leg movements to control individual avatar leg stepping.
Note: This is the specific experiment being extracted. Upright posture adds postural control demands compared to seated condition.
“identical interaction with the same virtual environment and BMI protocol while patients were upright, supported by a stand-in-table device”
Patient trains on Lokomat robotic gait trainer (Hocoma AG, Switzerland) which provides body weight support integrated with treadmill. Lokomat provides robotic-assisted gait training. Patient receives tactile feedback from Lokomat via haptic display on forearms synchronized with robotic foot rolling.
Note: This is component 3 of the WA-NR protocol. No BMI control in this component.
“training on a robotic body weight support (BWS) gait system on a treadmill (Lokomat, Hocoma AG, Switzerland)”
Patient trains with ZeroG body weight support system (Aretech LLC., Ashburn, VA) fixed on overground track. System rides along overhead fixed track with no mechanical barriers between patient and physical therapist. This setup requires patients to manage postural and trunk control, upper limb strength, and dynamic balance.
Note: This is component 4 of the WA-NR protocol. No BMI control in this component. More challenging than Lokomat due to lack of mechanical barriers.
“training with a BWS gait system fixed on an overground track (ZeroG, Aretech LLC., Ashburn, VA)”
Patient uses brain-controlled Lokomat robotic body weight support gait system on treadmill. Patient controls robotic gait via EEG-based BMI (leg imagination). Patient receives tactile feedback from Lokomat via haptic display on forearms synchronized with robotic foot rolling.
Note: This is component 5 of the WA-NR protocol. Combines BMI control with Lokomat robotic assistance.
“training with a brain-controlled robotic BWS gait system on a treadmill”
Patient uses custom-built 12 degrees of freedom robotic exoskeleton with brain control via EEG-based BMI (leg imagination). Exoskeleton has autonomous power, self-stabilization, and full lower limb hydraulic actuation. Patient is supported by ZeroG overground body weight support system with overhead track. Patient receives tactile feedback from exoskeleton via haptic display on forearms synchronized with robotic foot rolling. No crutches required.
Note: This is component 6 of the WA-NR protocol. Most challenging setup requiring postural control, trunk control, upper limb strength, and dynamic balance. Exoskeleton accommodates weight range 50-80 kg.
“gait training with a brain-controlled, sensorized 12 degrees of freedom robotic exoskeleton”
Training complexity increased over time to ensure cardiovascular stability and better postural control. Training progression starts with orthostatic training at stand-in-table and progresses to different gait training robotic systems.
Note: Complexity progression is individualized based on patient response
“the complexity of activities was increased over time to ensure cardiovascular system stability and better patient postural control; starting with orthostatic training at a stand-in-table and progressing all the way to the different gait training robotic systems”
Patient performs gait training using lower limb orthosis and walking assistive devices including hip-knee-ankle-foot orthosis or ankle-foot orthosis with knee extension splint and wheeled triangular walker.
Note: Further gait training component performed throughout protocol
“Further gait training was performed by having subjects utilize a lower limb orthosis and walking assistive devices (hip-knee-ankle-foot orthosis or ankle-foot orthosis with knee extension splint and wheeled triangular walker)”
Repeat comprehensive clinical assessments performed at 4, 7, 10, and 12 months using same battery as baseline: ASIA Impairment Scale, Semmes-Weinstein Monofilament Test, temperature/vibration/proprioception/deep pressure sensitivity evaluation, muscle strength test (Lokomat L-force Evaluation), Thoracic-Lumbar Scale, WISCI, SCIM, McGill Pain Questionnaire, Visual Analogue Scale for pain, range of motion assessment, Modified Ashworth Scale, Lokomat L-stiff Evaluation for spasticity, WHOQoL-Bref, Rosenberg Self-Esteem Scale, and Beck Depression Inventory.
Note: Routine general clinical evaluations also performed before and after every activity
“Such clinical evaluation started on the first day patients began training (Day 0), and were repeated after 4, 7, 10, and 12 months”
Throughout training, EEG signals recorded from 11 scalp electrodes over leg primary somatosensory and motor cortical areas while patients imagine movements of their own legs. Recordings performed before and after many months of training to evaluate potential functional cortical plasticity.
Note: Used to detect functional changes in leg representations over time
“patients were instructed to imagine movements of their own legs while EEG signals from 11 scalp electrodes were recorded over the leg primary somatosensory and motor cortical areas”
Independent Component Analysis employed to determine potential cortical sources (individual independent components) of novel leg representations in primary motor and somatosensory cortices and to detect functional changes of these representations over time.
Note: ICA used to identify cortical plasticity markers
“Independent Component Analysis (ICA) was employed to determine potential cortical sources, represented by individual independent components (ICs), of novel leg representations in the primary motor and somatosensory cortices”
For each independent component, Event Related Spectral Perturbations calculated with respect to baseline of 1 second prior to event and normalized by average power across trials at each frequency. Performed before and after many months of training.
Note: Used to evaluate brain dynamics modulation
“we calculated for each IC the Event Related Spectral Perturbations (ERSPs) with respect to a baseline of 1 second prior to the event and normalized by the average power across trials at each frequency”
Event Related Potentials sampled from two EEG electrodes located over leg representation area, averaged over all patients, before and after training. Used for statistical comparison of brain activity changes.
Note: Provides temporal information about cortical responses
“Event Related Potentials (ERPs), sampled from two EEG electrodes located over the leg representation area, averaged over all patients, before and after training, were also calculated and used for statistical comparison”
Paraplegic patients with chronic spinal cord injury (>1 year duration); seven complete and one incomplete SCI; all participants signed written informed consent