Brain Interfaces: Has the evolution only just begun?
Is it just hype or are brain interfaces here to stay? Are we at an inflection point? There has been an explosion of non-invasive modalities for the brain while at the same time a thrust to commercialize brain implant technology. At this point, it seems donning a ball-cap or helmet to read signals from the brain seems more accepting than a surgical implant. On another note, a brain implant to treat a chronic medical condition seems more acceptable while the idea of a brain implant to augment human performance leads to discussions toward fears of the unknown or potential unethical practices. Overall, are brain interfaces the right thing to do? Setting aside the technical and scientific jargon, our focus is to probe this area of neurotechnology. In this series, we will explore the origins of brain interfaces, embodiment options, the bionic pioneers of our time, and where we are headed in the future. If there is a specific area that you would like to see covered, post a comment and we will add it.
Brain Computer, Brain Machine or just Brain Interfaces
Human and machine interfaces have been evolving for several decades. The heart pacemaker has advanced tremendously since its first in human implant in 1958. Today, they seem to be commonplace with new bells and whistles like wireless communications, rechargeable batteries, and remote monitoring. Still, the heart pacemaker seems to be in a different class since it is an intervention for a muscular organ rather than the complex neural network of the brain.
“Your own mind is a sacred enclosure into which nothing harmful can enter except by your permission.” — Arnold Bennett
The brain seems to be this “sacred” aspect of the human body. Integrating technology with the brain in the form of an implanted device is available today but mainly for the treatment of medical conditions; such as responsive stimulation for epileptic seizure management or deep brain stimulation to settle tremors from Parkinson’s disease. These brain interface devices are mainly deployed within the medical model for the treatment of a specific condition. They have slowly gained acceptance over the decades of clinical deployment. On the other hand, when we begin to explore the potential for brain interfaces for purposes other than a medical treatment such as alternative communications, performance augmentation or cognitive enhancement, these concepts start to conger up debates within the public eye.
Brain waves are really not waves but electrical activity between neurons, called action potentials, to carry out functions in the brain.
Levels of Brain Interfaces
Brain interfaces are dependent on key technology components, specifically the ability to sense or listen to cortical signals. This ‘receiver’ of signals can detect small and diffuse neural activity in the brain and then interact with external controls. Using complex algorithms to ‘interpret’ the activity, monitoring systems that capture action potentials read like alphabet soup with an array of acronyms like EEG, MEG, and ECoG. The sophistication of the interfaces is wide ranging from a few signals to control a switch to complex webs of electrodes with high fidelity signaling.
Electroencephalography (EEG) are recording systems that process analog data coming from multiple electrode locations on the scalp. Whereas, Magnetoencephalography (MEG) systems monitor the magnetic signals emanating from the brain. The use of EEG and MEG in humans began in the early 1920s. The common method to capture these signals is the use of pasty electrodes placed on the scalp either individually or with a headset. More recently, alternative wearables like a net, baseball-type cap or even headsets with integrated sensors may be worn on the head to capture neural activity. Signals are captured and transmitted to the computer for further analysis. Advancements in the headgear reduce the use of conductive gels and conform to the diversity of individual anatomies. These neural activity capturing techniques are non-invasive, less expensive and quick to set-up and use. However, they lack the ability to decipher the activity on a nano level and implanted devices offer the needed specificity for complex tasks.
Electrocorticography (ECoG) are sensing systems using electrodes implanted directly to the subdural or dural surface of the brain to capture electrical activity. ECoG is widely used in clinical practice for people with epileptic zones in the brain and is considered the “gold standard” for this application. This technology has also been used for brain interface research to elicit external controls like a computer, robotic arm or neural prosthesis. Using implanted electrodes, ECoG can provide higher spatial resolution, better signal-to-noise ratio, and wider frequency range than its less invasive sensing counterparts. Advancements in materials used for implanted electrodes are of soft “silk” and polyimide to reduce the risk of potential brain tissue damage and eases the surgical implanting process.
Backed by a Rich History in Research
The science behind brain interface technology has a rich history of development. Going all the way back nearly 100 years ago, Han Berger demonstrated the use of EEG signals to detect brain activity in 1924. Fast forward to the late 1960s when Eberhard Fetz from the University of Washington discovered the operant conditioning of neural activity in non-human primates. Also in this era, Jon Wolpaw at the Wadswoth Center in New York developed the BCI 2000 using external EEG signals to interact with a computer which was later amended for home use. These early developments took the form of external devices or early wearables.
These modalities work well for simple tasks but they lack the ability to gain high bandwidth signaling. It’s like having dial-up internet access which worked for simple communication over the web but that same dial-up access does not have the capability to stream high-quality video.
It was the introduction of the Utah Array in the late 1980s that helped to marshal the early discoveries in implanted brain interfaces. The invention led by Richard Normann from the University of Utah was the gold standard electrode for early pre-clinical work in academic research institutions like Brown University (Donoghue), University of Washington (Fetz), University of Pittsburg (Schwartz/Gaunt), Johns Hopkins (Georgopoulos), Duke University (Nicolelis) and Northwestern (Miller). The discoveries in this era brought proof of concept to read brain neural activity to calibrate complex movements, communication and processing.
The leap of implanted brain technology occurred in 1998 when Philip Kennedy led the team to implant the first brain interface into a human in the form of the Neurotrophic electrode. The first pioneer was a person living with advanced ALS and the use case for the technology was to restore his ability to speak. Six years later in 2004, Matthew Nagle became this first person living with quadriplegia to be implanted with a brain interface allowing him to interact with a computer simply by thought. The electrode array was implanted into his motor cortex, the control center of the brain for movement, and the implant was connected to a computer. Other laboratories around the U.S. like Brown University (Donoghue/Hochberg), University of Pittsburgh (Schwartz/Gaunt), Caltech (Andresen), Utah (Florian), Stanford (Shenoy) and Battelle/The Ohio State University (Gaurav/Bouton) began the work of evolving the exploration toward the use of implanted brain interfaces to restore movement; for instance, using a robotic arm or electrical stimulation. The early bionic pioneers were people living with paralysis or amputation who were the brave souls to take the first steps names like Cathy Hutchinson, Jan Schumann, Ian Burkhart, Matthew Nagle, Nathan Copeland and Keven Walgamott to name a few. Today, there are approximately 30 people implanted with brain interfaces to control movement.
With a wide array of people being implanted, translational efforts are underway. In November 2014, the FDA hosted a workshop “Brain computer interface devices for patients with paralysis and amputation”. From this meeting and subsequent public input, the agency drafted the first guidance for BCI and regulatory evaluation. Much of this early work was supported by NIH Brain Initiative as well as various DARPA programs including RE-NET, RAM, RAM Replay, Neuro-FAST, NESD and N3.
To date there are many different types of external or wearable brain interface technologies on the market for simple applications however there are no commercialized implanted devices. At least, not yet.
Brain interfaces have drawn enough attention to attract some of the larger names in technology including Facebook and Google who are seeking to advance wearable technology, Bryan Johnson of Kernel to offer neuroscience/brain signaling as a service, and Elon Musk of Neuralink with a vision to make high fidelity high bandwidth brain implants as common as Lasik surgery.
Look for the next installment of this series featuring the array of brain interface embodiment options.
More information about the neurotech devices for various neurological conditions and other network resources may be found on the Neurotech Network website. Follow us on Twitter or LinkedIn.
For a easy to understand video series of BCI, visit BCI Guys on YouTube.
