Trial Review

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12621000779875
Ethics application status
Approved
Date submitted
31/03/2021
Date registered
22/06/2021
Date last updated
22/06/2021
Date data sharing statement initially provided
22/06/2021
Type of registration
Retrospectively registered

Titles & IDs
Public title
Can neurofeedback recover hand movement after stroke?
Scientific title
Can Restorative Brain-Computer interfaces Improve Hand Motor Functions after a Stroke?
Secondary ID [1] 303836 0
Nil known
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Stroke 321364 0
Condition category
Condition code
Stroke 319144 319144 0 0
Haemorrhagic
Stroke 319145 319145 0 0
Ischaemic

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
Can neurofeedback recover hand movement after stroke?


The rationale for Clinical Investigation: Finding novel and efficacious methods for motor recovery after stroke is a prominent unmet need. RehabSwift’s single-case study provided promising and clinically significant outcomes. Therefore, running a study involving up to 15 chronic stroke patients will determine whether and to what extent the observed results from the single case study can be generalized to the larger patient populations

Investigational Device:
The investigational device includes:
o an Electroencephalogram (EEG) cap and amplifier;
o a machine learning algorithm residing in a PC; and
o a bionic hand that provides haptic feedback.
o Electromyogram (EMG) electrodes attached to finger flexor/ extensor muscles.
The sponsor will provide the devices to the investigation site for the duration of the trial.

Study Procedures
Participants first attend a screening session during which they are screened for compatibility with the inclusion criteria. This includes monitoring patients vs the eligibility criteria elborated in step 5. Participants will also perform a brief run of motor imagery whilst their EEG signals are recorded. The overall screening session is expected to take up to two hours.
Those who pass the inclusion criteria, attend 18 neurofeedback training sessions over six consecutive weeks (three days per week). During training sessions, the performance of motor imagery of finger extension/flexion will be rewarded by the actual finger extension/flexion using bionic hands involved with participants’ affected hands.
Every neurofeedback training session starts with the preparation of the patient to wear an EEG cap and also attachment of up to four EMG electrodes to their arm muscles. Next, a measurement of the reaction time of the patient will be implemented.

Reaction time measurement will occur by involving the participant's hands in the bionic hands and having them following through with an extension or flexion movement of their fingers as fast as possible in response to a timed extension/flexion movement from the bionic hands. The response between the bionic hand movement and participant muscular activity will be monitored.

Then eight runs of neurofeedback training will be implemented, where each run includes conducting 20 motor imagery and relaxation trials in a randomised order. The motor imagery trials consist of asking the participants to imagine extend/flex their fingers. Relaxation trials will involve asking participants to remain still and do nothing such as concentrating on a blank wall. There will be 2-minute breaks between consecutive runs and collectively each neurofeedback training session is expected to take approximately one hour.

Participants' attendance will be recorded as well as lab notes corresponding to each session.

Neurofeedback training sessions will be provided by an experienced and well-trained experimenter with a background in neuroscience and/or psychology. The intervention will be provided face to face and individually. The intervention would take place at the University of Adelaide where all the necessary machinery will be provided by the sponsor for the duration of the trial.
Intervention code [1] 320140 0
Rehabilitation
Intervention code [2] 320141 0
Treatment: Devices
Comparator / control treatment
No control group
Control group
Uncontrolled

Outcomes
Primary outcome [1] 327023 0
Fugl-Meyer Upper Extremity Test (FME-UE)

The Fugl-Meyer is a scale that measures capacity of the arm. There are 33 items that evaluate reflexes, active movement, grasp patterns, speed, and coordination of the arm and hand. Each item is scored from 0-2, with a maximum total score of 66. Higher scores indicate better performance.
Timepoint [1] 327023 0
Completed at three time points;
-Prior to intervention
- At the Conclusion of intervention
-four to six weeks post conclusion of the intervention
Secondary outcome [1] 393485 0
Action Research Arm Test (ARAT)
The ARAT is a scale that measures activity participation of the arm. There are 19 items that several types of grasps, grips, pinches, and gross movements of the arm and hand. Each item is scored from 0-3, with a maximum total score of 57. Higher scores indicate better performance.
Timepoint [1] 393485 0
Completed at three time points;
-Prior to the intervention
- At the conclusion of the intervention
-four to six weeks post conclusion of the intervention
Secondary outcome [2] 393488 0
Goal Attainment Scale (GAS)
Timepoint [2] 393488 0
Completed at three time points;
-Prior to the intervention
- At the conclusion of the intervention
-four to six weeks post conclusion of the intervention
Secondary outcome [3] 393489 0
Modified Ashworth Scale

The Modified Ashworth scale measures the resistance to passive movement about a joint with varying degrees of velocity. Scores range from 0-4 with a 1+ scoring category to indicate resistance through less than half the movement, with 6 choices.
A score of 0 indicates no increase in muscle tone and a score of 4 indicates the affected part is rigid in flexion or extension.
For this study, only fingers and wrist extension, and flexion were monitored.
Timepoint [3] 393489 0
Completed at three time points;
-Prior to the intervention
- At the conclusion of the intervention
-four to six weeks post conclusion of the intervention
Secondary outcome [4] 393490 0
Pinch and Grip test
Timepoint [4] 393490 0
Completed at three time points;
-Prior to the intervention
- At the conclusion of the intervention
-four to six weeks post conclusion of the intervention
Secondary outcome [5] 393491 0
Reaction Time
The reaction time is measured by involving participants' hands in the bionic hands and having them following through with an extension or flexion of their fingers as fast as possible in response to extension/flexion movement from the bionic hands. During this test, the electromyogram (EMG) signal of the finger flexor or extensor of the arm muscle are also recorded. The response time is measured by a built-in algorithm that continuously monitors changes of the recorded EMG signals.
Timepoint [5] 393491 0
Completed prior to each training session.
Secondary outcome [6] 393492 0
Neglect test
Behavioural assessment of unilateral neglect (https://pubmed.ncbi.nlm.nih.gov/12589620/)
The test involves asking the participants 10 questions, where each question gets a mark between 0-3. The higher the number, the more behavioural neglect is experienced by the participant.
Timepoint [6] 393492 0
Completed at three time points;
-Prior to the intervention
- At the conclusion of the intervention
-four to six weeks post conclusion of the intervention

Eligibility
Key inclusion criteria
The prospective participants have to fulfill the following inclusion criteria:
1- being at least six months post-stroke and in a stable condition;
2- having impaired motor capabilities in their affected arm determined by an ARAT score less than 45 out of 57;
3- having intact cognitive functions determined by the mini-mental state examination (MMSE) score to be more than 26 out of 30;
4- being independently mobile—with or without a walking aid;
5- not having excessive tone in their arm and hand muscles determined by the modified Ashworth test score to be less than 3 out of 4;
6- having the ability to perform vivid MI—by screening their ability in generating discriminable MI vs relaxation EEG signals;
7- having an (almost) intact sense of proprioception—by screening their blind judgment of comparing the size of seven polystyrene balls with more than 50% accuracy;
8- the ability to fully understand and comprehend auditory instructions presented in English to perform motor imagery.
Minimum age
18 Years
Maximum age
80 Years
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
1- individuals with comorbidities such as arthritis in the hands/fingers of their affected side will be excluded.
2- individuals who do not think they can fulfil the visit attendance requirements will be excluded.
3- Visual or hearing impairment.

Study design
Purpose of the study
Treatment
Allocation to intervention
Non-randomised trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Masking / blinding
Who is / are masked / blinded?



Intervention assignment
Other design features
Phase
Not Applicable
Type of endpoint/s
Safety/efficacy
Statistical methods / analysis
Description of Statistical Methods
There is a consensus (https://academic.oup.com/ptj/article/92/6/791/2735128) that for chronic stroke survivors, a minimum increase of 5.25 in their Fugl-Meyer upper extremity score is required for the therapy to be considered as clinically important. Therefore, the primary clinically meaningful effect of the therapy was determined to be an increase of 5.25 points in Fugl-Meyer upper extremity scores; the standard deviation was assumed to be 4 points based on evidence from the available literature. In the absence of information concerning the magnitude of the correlation between pre- and post-treatment scores, a worst-case scenario was adopted whereby the correlation was set to zero. Under these assumptions, a sample of 12 patients would be required for the study.

The Number of Participants
The recruitment process and the main study will run in parallel. In case of drop outs recruitment is continued till the target sample size of 12 is reached.

The Level of Statistical Significance
Power calculations were based on the requirement that effects be assessed at the 5% alpha-level with 80% statistical power.

Recruitment
Recruitment status
Recruiting
Date of first participant enrolment
Anticipated
Actual
11/11/2020
Date of last participant enrolment
Anticipated
15/09/2021
Actual
Date of last data collection
Anticipated
24/12/2021
Actual
Sample size
Target
12
Accrual to date
8
Final
Recruitment in Australia
Recruitment state(s)
SA
Recruitment hospital [1] 19043 0
The Queen Elizabeth Hospital - Woodville
Recruitment postcode(s) [1] 33590 0
5011 - Woodville

Funding & Sponsors
Funding source category [1] 308231 0
Commercial sector/Industry
Name [1] 308231 0
RehabSwift Pty ltd.
Country [1] 308231 0
Australia
Primary sponsor type
Commercial sector/Industry
Name
RehabSwift
Address
10 Pulteney St, Adelaide, SA, 5000
Country
Australia
Secondary sponsor category [1] 309027 0
None
Name [1] 309027 0
Address [1] 309027 0
Country [1] 309027 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 308213 0
The Univeristy of Adelaide
Ethics committee address [1] 308213 0
The University of Adelaide
Adelaide, South Australia
5005 Australia
Ethics committee country [1] 308213 0
Australia
Date submitted for ethics approval [1] 308213 0
30/06/2020
Approval date [1] 308213 0
05/08/2020
Ethics approval number [1] 308213 0
H-2020-147
Ethics committee name [2] 308216 0
Central Adelaide Local Health Network (CALHN)
Ethics committee address [2] 308216 0
Central Adelaide Local Health Network
Human Research Ethics Committee
North Terrace
Adelaide, SA, 5000
Ethics committee country [2] 308216 0
Australia
Date submitted for ethics approval [2] 308216 0
24/08/2020
Approval date [2] 308216 0
27/01/2021
Ethics approval number [2] 308216 0
2020/HRE00315 (formally 13831)

Summary
Brief summary
Stroke survivors suffer from impaired movements that affect their activities of daily living. Conventional physio and occupational therapies are provided to regain some of these lost motor abilities. However, even with intensive conventional therapy, a large portion of patients do not receive sufficient recovery to function independently. This deficit leads to significant social and economic burdens. Therefore, there is a clear need for the development of novel and effective intervention strategies. To address the gap, alternative therapies, including brain-computer interfaces, have been proposed to recover impaired hand functions following stroke.
A brain-computer interface (BCI) can translate the imagination of hand movement to the movement of an object on a screen or even the actual hand movement provided by a robotic hand. In a BCI system, a specific cap records the electrical activity of the brain on the surface of the scalp during the imagination of the hand movement. A machine learning algorithm processes the recorded brain signals and determines if the imagination of movement has been performed. When the algorithm detects the movement imagination, it sends commands to the outside world and provides feedback to the user. A specific type of feedback may be supplied via a robotic hand to move a paralysed hand passively, during the imagination of the hand movement. There is a growing body of research, demonstrating the promising primary results of the application of BCIs for movement recovery after stroke. However, to supply BCI as standard therapy in clinics, its consistency and efficacy need to be improved.
One of our prior studies suggests the outperformance of proprioceptive feedback over traditional visual feedback in the provision of a better substrate for the occurrence of operant learning. Further, we have demonstrated that provision of the correct delay between the brain activation and the passive hand movement is critical and must be customised according to individual’s attributes. Putting together the findings of the two prior studies, we improved the hand movement of a stroke patient, who had had a stroke 3.5 years before the study, by 36% after ten sessions of therapy.
Observing the promising results of the aforementioned studies, we are investigating how our novel BCI therapy recovers the hand movement for a group of stroke survivors.
Trial website
https://www.rehabswift.com/clinical-trials
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 109954 0
Prof Derek Abbott
Address 109954 0
Room 3.47 , Ingkarni Wardli , North Terrace, The University of Adelaide, Adelaide SA 5000
Country 109954 0
Australia
Phone 109954 0
+61883135748
Fax 109954 0
Email 109954 0
[email protected]
Contact person for public queries
Name 109955 0
Dr Dr Sam Darvishi
Address 109955 0
RehabSwift
10 Pulteney Street Adelaide SA 5000
Country 109955 0
Australia
Phone 109955 0
+61450214545
Fax 109955 0
Email 109955 0
[email protected]
Contact person for scientific queries
Name 109956 0
Prof Derek Abbott
Address 109956 0
Room 3.47 , Ingkarni Wardli , North Terrace, The University of Adelaide, Adelaide SA 5000
Country 109956 0
Australia
Phone 109956 0
+61883135748
Fax 109956 0
Email 109956 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment


What supporting documents are/will be available?

No Supporting Document Provided



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.