Brain Machine Interfaces based on Subcortical LFP Signals for Neuroprosthetic Control and Neurofeedback Therapy

Lead Research Organisation: University of Oxford
Department Name: Clinical Neurosciences

Abstract

Recovering upper limb function will offer a certain degree of independence and sense of autonomy to people with paralysis due to disabling spinal cord injury, amputation, stroke etc. The 'Brain machine interfaces' (BMIs) convert brain signals into control signals for guiding prosthetic arms or other devices, and have showed great potential to restore functions important for everyday life, such as reaching and grasping. However, the translation of the exciting research progress to clinical use that actually improves the daily lives of people with disabilities has barely begun. Key challenges in the clinical applications of existing BMIs include: 1) Difficulties in ensuring stable and satisfactory recordings of brain signals over months or years. Loss of signals over time leads to deterioration in the performance of the neuroprosthetic device and frustration in users. 2) The difficulty in accurately and reliably estimating certain movement parameters such as the gripping force in a simple grasp movement. To date, the best clinical demonstration of BMI has still not been able to accurately manipulate the force level that was applied by a robotic hand.

Research into BMI has, to date, almost exclusively focused on signals obtained from the surface or the upper layer of the brain (the cerebral cortex). My previous research has identified the important role of brain signals from the 'basal ganglia', a structure deep inside the brain, in controlling movement and representing gripping force in a grasp. These signals can be readily recorded from electrodes that last many years and such electrodes can be implanted in a relatively safe procedure, which has been a routine therapy for movement disorders. Therefore, these signals offer significant advantages for long-term performance and reliability of the BMI over time. I propose using these deep brain signals to control the grip force of robotic hands, and to study how the patients learn to use the prosthetic hand. This will provide important proof-of-principle of using signal recorded from structure deep inside the brain to control a robotic hand. Meanwhile, understanding and engaging the process of BMI skill learning will potentially be another major opportunity for further improvement of the performance of BMIs.

Importantly, what is central to BMI use is for a subject to achieve a specific goal by voluntarily changing their brain activity. But could the BMI be used to train patients to change their own pathological brain activity that is causing problems? Positive answer to this question can lead to novel therapies to diseases where a clear pathological brain signal has been identified. For example, pathological brain activity in the basal ganglia has been heavily associated with motor impairment in a range of diseases, such as Parkinson's disease (PD). I will use the BMI system proposed here to train patients with PD to reduce the pathological signals in the targeted brain area while giving them the feedback about the level of this pathological signal (so called 'neurofeedback training'). I will test the hypothesis that when given feedback, patients are able to reduce the pathological signal that is causing problem in their disease, and that voluntary change of the pathological activity can lead to improvement in movement related symptoms in PD. This work will also help to shed light on the underlying mechanisms of neurofeedback training, which may facilitate other effective clinical applications of this technique.

In summary, this work will establish the foundations for novel brain-machine interfaces based on signals recorded from deep brain regions that contain rich information related to movement intention and have been proven to be stable over time. I will use the new framework to control a prosthetics hand with graded gripping force, to provide neurofeedback training to reduce symptoms in PD, and to study the role of basal ganglia in the control and learning of movements.

Technical Summary

Research into BMI has, to date, almost exclusively focused on signals obtained from the cerebral cortex. The clinical application of BMI faces grand challenges including the lack of longevity in the neural spiking recorded over the cortex for decoding movement related information, and the difficulty in reliable and accurate prediction of gripping force required to develop neuroprosthetics of clinical grade.

My previous research has identified the important role of local field potentials (LFP) recorded from the basal ganglia in encoding movement related information such as motor effort and gripping force, suggesting that these signals can be used to deliver graded force generation via a robotic hand. Compared to neural spikes, the long-term stability of LFP signals offers significant advantages for long-term performance and reliability of the BMI over time. In addition, new developments in implantable devices, which offer the capacity for chronic recording, real-time processing and wireless data transmission, afford an unparalleled opportunity to utilize deep brain signals for BMI control.

I propose to exploit this research opportunity and our understanding about the role of basal ganglia in motor control, to develop new approaches for BMI based on LFP signals recorded from the basal ganglia. I will build the first robotic mechanism interfacing with subcortical structures - the basal ganglia, and use the new framework to control a prosthetics hand with graded gripping force. This will also allow me to investigate the neural basis of the learning process in the use of a neuroprosthetic hand in human subjects. Secondly, I will use the system to provide neurofeedback training to regulate pathological oscillations in the basal ganglia associated with Parksinson's disease, and to study the functional changes in the cortico-basal ganglia network induced by neurofeedback training.

Planned Impact

The research in BMI systems generates tremendous excitement from academic community, clinicians and general public. The excitement reflects the rich promise of BMIs. People who may benefit from the proposed research include:

1) Staff working on the project.
Through the project, the PI will further develop her long-term career plan and scientific vision to translate neurosciences and engineering knowledge into clinical therapies in order to recover function and ameliorate symptoms. The Research Associate will gain essential training in biomedical signal processing, real-time implementation of robotic control, experimental design and recording in clinical environment.

2) Healthcare industry.
The current proposal opens the door for new applications of clinical grade implantable pacemakers that also include ultra-low power amplifier and wireless telemetry for data uploads. More clinical use of the device is a new commercial opportunity for the healthcare industry.

3) People suffering from Parkinson's disease.
There are 127,000 people with Parkinson's disease in UK, and 10-20% of them will receive the surgery for deep brain stimulation therapy in the late stage of the disease. They will benefit from the neurofeedback training therapy proposed in the study, especially since DBS pacemakers with the capacity to measure and wirelessly transmit data are becoming widely used in clinical applications. Self-regulation of the pathological neural activities combined with potential plastic changes in the brain will reduce the requirement for medication, or high amplitude of DBS stimulation. This will reduce the side effects that are concomitant with both medication and DBS, and improve the quality of life of the patients.

4) People living with paralysis.
In the UK and Ireland, there estimated to be 50,000 people with paralysis due to spinal cord injury, amputation, stroke and etc. among which the spinal cord injury primarily affects young adults. The cost to the nation is estimated at £1 billion per annum. Recovering upper limb functions will offer both functional independence, and a sense of autonomy, which is of substantial benefit to these people, and will reduce social support required for them. At present, the achievements of BMI research and development remain confined almost entirely to the laboratory. The proposed research will make significant contributions towards making the neural prostheses more reliable in long-term.

5) People suffering from other disease in which neurofeedback therapy may help.
There is increasing number of patients with psychological disorders such as anxiety, depression and post-traumatic stress disorder, who are referred to non-invasive neurofeedback training. Neurofeedback training has also been proposed to improve rehabilitation for people with strokes, head trauma, and other disorders. My study on the potential plastic changes in the larger network of the brain induced by the neurofeedback training will help to shed light on the underlying mechanisms of neurofeedback training that will undoubtedly facilitate more effective clinical applications.

6) Wider public.
There has been widespread and lasting societal interest in the development and research on BMI systems. This is possibly because BMIs not only have the potential to revolutionize the life of many people with motor disabilities, but also revolutionize the way people communicate with the world, interact with information technology and the world itself. Outlined research can benefit the wider public by increasing the awareness of the existence of the technique, and by providing insights into how BMIs can contribute to improving people's life in multiple ways.

Publications

10 25 50
publication icon
Torrecillos F (2018) Modulation of Beta Bursts in the Subthalamic Nucleus Predicts Motor Performance. in The Journal of neuroscience : the official journal of the Society for Neuroscience

 
Description Rosetrees Trust Research Grant
Amount £60,000 (GBP)
Organisation Rosetrees Trust 
Sector Charity/Non Profit
Country United Kingdom
Start 04/2018 
End 03/2021
 
Description Wellcome ISSF
Amount £59,880 (GBP)
Organisation Wellcome Trust 
Sector Charity/Non Profit
Country United Kingdom
Start 04/2018 
End 03/2019
 
Title Machine learning based methods for decoding force based on STN LFPs 
Description We have developed a few methods, such as a linear dynamic model and a Wiener cascade model, to decode gripping force based on STN LFPs. 
Type Of Material Model of mechanisms or symptoms - human 
Year Produced 2017 
Provided To Others? Yes  
Impact On one hand, our approach has offered new insight on how basal ganglia is involved in controlling gripping force, and also offer new understanding about the pathology of Parkinson's disease. This has led to further research in other groups citing our work. On the other hand, the method also pave the way for further development of the Brain Machine Interfaces based on subcortical signals, which is the main goal of the funded project. 
 
Title STN LFPs and simultaneous EEG during finger joystick movement 
Description These data contain local field potentials (LFPs) from the Subthalamic Nucleus and simultaneous EEGs recorded from patients with Parkinson's Disease while they performed a finger joystick motor task after they received surgery for Deep Brain Stimulation. The analysis of data was first presented in the paper Torrecillos et al (2018) where the details of the experimental design and behaviour task are presented. All EEGs and LFPs were recorded with common reference rejection and the ground was attached to the wrist of the patient. 
Type Of Material Database/Collection of data 
Year Produced 2018 
Provided To Others? Yes  
Impact Thanks to the close collaboration we have established with functional neurosurgery teams, we have enjoyed the rare opportunity to record local field potentials from the deep brain structures, such as the basal ganglia, in human patients with movement disorders who have undergone the surgery for Deep Brain Stimulation. We will continue to freely share those data in order to support open science and reproducibility in neuroscience. 
URL https://data.mrc.ox.ac.uk/data-set/simultaneous-stn-lfps-and-eegs-human-patients-during-finger-joyst...
 
Title STN LFPs from Parkinson's patients during stepping in place 
Description These data contain local field potentials (LFPs) from the human subthalamic nucleus recorded from patients with Parkinson's after receiving deep brain stimulation surgery. Patients were asked to step on the spot (while sitting and for three data sets also while standing) and to synchronize their steps to the rhythm of a walking cartoon man displayed in a video. The details of the experimental design and behavioural task are described in Fischer et al (2018). Results on decoding analyses, attempting to decode movement states within the gait cycle based on the STN LFPs, are reported in Tan et al (2018). All LFPs were first recorded as monopolar signals with a common reference and the ground electrode attached to the wrist of the patient. Bipolar LFPs were constructed by computing the difference between monopolor recordings from neighbouring contacts. The timing of each heel strike was simultaneously recorded with foot pedals or force plates. 
Type Of Material Database/Collection of data 
Year Produced 2018 
Provided To Others? Yes  
Impact Thanks to the close collaboration we have established with functional neurosurgery teams, we have enjoyed the rare opportunity to record local field potentials from the deep brain structures, such as the basal ganglia, in human patients with movement disorders who have undergone the surgery for Deep Brain Stimulation. We will continue to freely share those data in order to support open science and reproducibility in neuroscience. 
URL https://data.mrc.ox.ac.uk/data-set/stn-lfps-parkinsons-patients-during-stepping-place
 
Title Subthalamic nucleus activity during stopping of rhythmic finger tapping 
Description Through close collaboration with functional neurosurgery teams throughout UK, we have enjoyed the precious opportunity to record brain activities from deep structures of the brain in patients with Parkinson's Disease, who received the surgery for Deep Brain Stimulation as a treatment for their diseases. We make the data we collected freely available to other scientist through Oxford University Research Archive. The data was collected between 2015 and 2016 at the John Radcliffe Hospital (Oxford) and the National Hospital for Neurology and Neurosurgery (London) in 9 patients with Parkinson's disease who had undergone deep brain stimulation surgery. Patients performed a stopping task while local field potentials from the subthalamic nucleus and scalp EEG channels (C3, C4, Cz, Fz, Pz) were recorded with a TMSi Porti amplifier. Patients were instructed to interrupt finger tapping, which was paced by a metronome, in response to an auditory stop-signal, which was timed such that successful stopping would occur only in ~50% of all trials. This data will allow us to better understand how activities from different areas of the brain are involved in rhythmic movements and the successful stopping of those ongoing movements. 
Type Of Material Database/Collection of data 
Year Produced 2017 
Provided To Others? Yes  
Impact The dataset has been downloaded for 40 times since it was made available online in 2018. 
URL https://ora.ox.ac.uk/objects/uuid:54c00c3d-1809-4a52-bba8-b491b6075f35
 
Description Clinical collaborator in King's College Hospital 
Organisation King's College Hospital
Country United Kingdom 
Sector Hospitals 
PI Contribution The research team design and carry out the research study.
Collaborator Contribution The partner is offering clinical care to potential participants and help to identify and recruit participants for the study.
Impact Four participants have been recruited through the collaboration.
Start Year 2017
 
Description Clinical collaborator in Oxford 
Organisation University of Oxford
Country United Kingdom 
Sector Academic/University 
PI Contribution The research team will design and carry out the research study.
Collaborator Contribution The clinical team and neurosurgeons will be offering clinical care and help to identify potential participants for the study.
Impact Three participants have been recruited through the collaboration.
Start Year 2017
 
Description Clinical collaborator in UCLH 
Organisation University College London Hospital
Department University College London Hospitals Charity (UCLH)
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution The research team design and carry out the study.
Collaborator Contribution The clinical partner offer clinical care to the patients and help to identify and recruit participants in the study.
Impact Two participants have been recruited through the collaboration.
Start Year 2017
 
Description Collaborator in Marseille 
Organisation Institute of Neuroscience of Timone
PI Contribution With my expertise in electrophysiology and in Parkinson's disease, I have contributed to coming up with a new research proposal for a grant application together with my collaborator. I will also help to train staff.
Collaborator Contribution Through my partner, the collaboration will allow me to extend my research, to test different research ideas and to have access to patients in Marseilles.
Impact A new proposal has been submitted.
Start Year 2017
 
Description Fudan University (Neural and Intelligence Engineering Center) 
Organisation Fudan University
Country China 
Sector Academic/University 
PI Contribution I visited the partners in Fudan University and shared with them some of my research ideas, and propose to analyse some data which are available together from a different perspective.
Collaborator Contribution The partners are in Neural and Intelligence Engineering Center, Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai, China. The partner have provided data they previously collected which is very relevant to my project: decoding voluntary movements and postural tremor based on thalamic LFPs for closed-loop stimulation for essential tremor.
Impact I analysed the data and a manuscript has been submitted to Brain Stimulation and currently under review.
Start Year 2018
 
Description Oxford OHBA 
Organisation University of Oxford
Department Oxford Centre for Human Brain Activity (OHBA)
Country United Kingdom 
Sector Academic/University 
PI Contribution My team proposed to apply the methods developed by the collaborators on LFPs recorded from basal ganglia in order to decode movement status for advanced BMIs.
Collaborator Contribution The partners have developed a method that can be used for the current project.
Impact A join publication has now been submitted, and new projects and research ideas have been developed.
Start Year 2017
 
Description University Hospital Cologne 
Organisation University Hospital Cologne International
Department Department of Functional Neurosurgery and Stereotaxy
Country Germany 
Sector Hospitals 
PI Contribution I contacted the partners, shared with them some of my research ideas, and propose to analyse some data which are available from their center from a different perspective.
Collaborator Contribution The partner have provided some precious data they previously collected which is very relevant to my project: decoding voluntary movements and postural tremor based on thalamic LFPs for closed-loop stimulation for essential tremor.
Impact I analysed the data and a manuscript has been submitted to Brain Stimulation and currently under review.
Start Year 2018
 
Title Measurement of an Electrophysiological Signals During Stimulation of a Target Area of a Body 
Description When generating a stimulation signal comprising stimulation pulses delivered to a target area of a human or animal body, an electrophysiological signal measured from the body for closed-loop control of the stimulation signal, is sampled, at a sampling frequency in an analogue-to-digital converter for deriving a feedback signal for closed-loop control of the stimulation signal. The generation of the stimulation signal and the sampling of the electrophysiological signal are synchronised and have a relative phase selected to cause the sampling to occur outside the stimulation pulses, which prevents the effect of the stimulation pulses from interfering with the digital electrophysiological signal, while allowing maintenance of Nyquist-Shannon rules and the integrity of the discrete Laplace transform (z-transform) required in discrete control theory. 
IP Reference UK Intellectual Property Office Patent Application Number 1816141.4 
Protection Patent application published
Year Protection Granted 2018
Licensed No
Impact Closed-loop DBS has shown great potential in reducing side effect in people with Parkinson's disease (PD); and closed-loop approaches are receiving increasing attention in 30 other fields such as deep brain stimulation for psychiatric disorders, spinal cord stimulation, peripheral or autonomic nerve stimulation and non-invasive brain stimulation. However, this new approach suffers from an important technological limitation that the electrophysiological signal measured close to the stimulation target as a feedback signal is corrupted by large electrical artefacts derived from the stimulation signal. This is an inevitable consequence of the electrical stimuli delivered simultaneously during recording. This leads to significant problems which corrupt the extracted LFP signal and making it less suitable for closed-loop applications. The invention that has been submitted for patent protection is a method to significant reduce stimulation artifact in the measurements of electrophysiological signals simultaneously recorded with stimulation. It will resolve a main obstacle for any type of closed-loop modulation that uses the electrophysiological signals recorded close to the stimulation site as a feedback signal in order to adapt the stimulation to the physiological states.
 
Description In2Science UK 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Schools
Results and Impact Five Year 13 students were enrolled on a 2-week programme (non-residential) during which are given 1.) personalised mentoring from scientists; 2.) Opportunities to gain a wide variety of practical experiences as well as exposure to key concepts and challenges in research; 3.) Integrated workshops with in2scienceUK, where the pupils receive guidance on university applications, wider information about STEM careers, and training in transferable skills. I give presentations to those students, show them the studies that we are conducting, and show them different research tools my team is using.
Year(s) Of Engagement Activity 2017
URL http://www.mrcbndu.ox.ac.uk/outreach
 
Description MRC Brain Network Dynamics Unit Open Day 
Form Of Engagement Activity Participation in an open day or visit at my research institution
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Schools
Results and Impact The MRC Unit open day aims to to encourage students into science and research, and to demonstrate the facilities within the Unit and the research being conducted. The open day has occurred every year. Each year, around 120 students/teachers from local schools will visit the unit. My team will have a demonstration about trans-cranial magnetic stimulation, showing how TMS can be used in research, as well as in clinical diagnosis and therapy. We conduct simple and fun experiments which can help to answer some questions about how the brain controls movements.
Year(s) Of Engagement Activity 2017,2018
 
Description Patient group presentation 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Patients, carers and/or patient groups
Results and Impact Together with a D.Phil. student from the MRC Brain Network Dynamics Unit at the University of Oxford, Eszter Kormann, I attended a branch meeting of the Parkinson's UK at Banbury on 17th Aug to visit about 40 members of a local group of people affected by Parkinson's, including patients, carers and their friends and families.

The visit began with a talk from Eszter on the role of brain rhythms in Parkinson's, drawing on her work with patients as well as the use of animal models in Parkinson's research. Huiling then introduced some of the Unit's research on brain stimulation for the treatment of Parkinson's, and shared some exciting new developments for improving these therapeutic approaches. Each of the talks was integrated with a lively discussion session in which the audience's questions came thick and fast, stimulating further conversations about Unit discoveries made in the clinic and at the lab bench.

Feedback from the audience was overwhelmingly positive, and included: "The presentations were pitched at just the right level for our members, very few of whom have any significant scientific or medical knowledge." "I was very interested and encouraged to learn that closed-loop Deep Brain Stimulation may be available to some Parkinson's patients in the future."
Year(s) Of Engagement Activity 2018
URL https://www.mrcbndu.ox.ac.uk/news/eszter-and-huiling-take-unit-science-out-local-people-affected-par...
 
Description Workshop Presentation 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact It is an international workshop organized by the Israel Science Foundation and the British Council. The workshop aims to bring together a multidisciplinary group of experimental and theoretical neuroscientists for a series of lectures and discussions, to encourage collaboration and widen the vision of undergraduate/postgraduate students in Israel.
Year(s) Of Engagement Activity 2017