From Movies to Reality: Controlling Devices With Our Minds

Executive Summary

• Two general classes of Brain-Computer-Interfaces (BCIs) include invasive and non-invasive.
• To date, invasive BCI focuses on being granted medical device status targeting individuals with neurological disorders with limited use for healthy individuals.
• Non-invasive BCI is the preferred approach for commercial use, but has not yet managed to scale. Until recently, the technology generally provided low signal quality, had high price points, impractical form factors, and did not offer a good user experience.
• With recent breakthroughs in form factor, accuracy of signal acquisition and processing, non-invasive BCIs will allow individuals to obtain real-time insights and behavioral impulses. These insights will enable people to be more focused, avoid burnout, lessen mental stresses, sleep better, and be more conscious of themselves.

Introduction to BCI

Some implantable BCI companies, such as Blackrock Neurotech, have proven that Brain-Computer-Interfaces (BCIs) can greatly support individuals who are severely disabled by disorders such as ALS, cerebral palsy, or other neurology-related conditions. BCIs enable users to operate external devices not controlled by peripheral nerves or muscles via brain activity through processed electrical signals.

The technology works by capturing signals (electrical or magnetic) from the brain, which obtains information contained therein. BCIs can thus serve as a replacement for normal neuromuscular pathways as it allows individuals to activate assistive technologies to enhance communication or control things. For example, by 2006, a microelectrode array was implanted in the primary motor cortex of a young man, and the signals obtained from there with a BCI system enabled him to operate a TV and perform actions with a robotic arm (1).

BCI technologies have developed further for more comprehensive applications and usage beyond clinical settings in recent years. Particularly in robotic and interactive systems, people can now mind-control a light switch or send a message to their parents by just thinking about it.

To understand where BCI technology for everyday life can go, we recommend watching this visionary video by Neurable (est. 2015, US).

Funding and TAM

The space has seen a sizable increase in the number of emerging start-ups, research and funding in recent years, with 2021 reaching $130mm of capital raised across 14 deals. Our study shows the current trajectory continuing into 2022 and beyond.

According to Allied Market Research, the BCI market (invasive and non-invasive) is valued at around $1.5bn and is expected to grow by almost 14% per year to $5.5bn by 2030.

Invasive (Implants) vs. Non-Invasive (Headsets)

Achievements in BCIs have relied on a mix of invasive and non-invasive technologies. BCIs that are invasive, using electrodes implanted in the brain, tend to deliver truly accurate signal measurements compared to most non-invasive options.

Invasive

Many neuroscientists rely on a surgically embedded array of electrodes, such as those produced by a Utah-based company like Blackrock Neurotech. Their invasive technology can differentiate the signals of individual neurons, allowing more refined control of connected devices. However, surgery can result in infection, inflammation, and scarring, which may eventually degrade signal strength. Still, a major recent breakthrough in the field was Synchron’s (est. 2016, US) May 4th announcement of the enrollment of the first patient in its US clinical trial, COMMAND, for patients with severe paralysis with a permanently implantable BCI (2). Previous BCI human clinical studies approved by FDA have been conducted in short-term experimental settings.

Their device, the Stentrode, is an endovascular implant, therefore minimally invasive, but at least it does not require open-brain surgery. The stentrode is four centimeters in length and is a lattice of electrodes. It is inserted via the jugular vein into one of the brain’s blood vessels and could change the lives of patients who have paralysis and other debilitating neurological diseases.

Synchron’s use case exemplifies many researchers and physicians’ predominant interest in developing BCIs for therapeutic applications, i.e., restoring movement and communication to people who are paralyzed or otherwise disabled. For such use cases, the highest levels of precision signal acquisition are required and usually obtained invasively. Yet beyond clinical applications, the apparent potential of such technology and the increasing number of high-profile start-ups developing BCI tech suggests the possibility of much wider adoption and a much larger TAM.

Non-invasive

A key benefit to non-invasive BCIs is that they have the potential to solve one of the most significant problems in neuroscience: access to large, varied, and longitudinal data sets. Other than for sleep studies, data sets for brain data are small: 1,000–2,000 hours of data for epilepsy and <100 hours for affective computing, BCIs, and cognitive data. A non-invasive everyday BCI will revolutionize neuroscience discovery and accelerate advanced BCI development.

For this realm of applications to become a reality, the accuracy and practicality of non-invasive BCIs will play an instrumental role. EEG-based non-invasive BCIs are currently the most widely researched approach owing to the minimal risk involved and the relative convenience of conducting studies and recruiting participants (3).

Clinical studies show that people with severe motor disabilities do not require electrodes implanted in their brains to control movement. In 2004, non-invasive BCIs using scalp-recording activity and an adaptive algorithm could provide humans with multidimensional movement control that falls within the range reported with invasive methods in monkeys (4).

This points towards the promising trajectory of enabling non-invasive BCI technology for commercial use.

Non-Invasive BCI Tech Infrastructure

The purpose of a BCI is to detect and quantify features of brain signals that indicate the user’s intentions and to translate these features in real-time into device commands that accomplish the user’s intent. To achieve this, a BCI system consists of 4 sequential components: 1) signal acquisition, 2) feature extraction, 3) feature translation, and 4) device output (2).

EEG, first used by Hans Berger in 1924, works by safely collecting electrical signals produced by your brain. EEG measures this electrical activity by recording voltage fluctuations due to the flow of ionic current during synaptic excitations in the brain’s neurons. Each of the billions of neurons in your brain produces brainwaves, and EEG sensors safely detect significant changes in brainwaves. In other words, EEG-based BCI is a direct connection between the human brain and the computer.

Neurofeedback then (or EEG feedback) teaches self-control of brain functions to subjects by measuring brain waves and providing a feedback signal, which can be issued as audio or video feedback. During this process, the subject becomes aware of the changes taking place and will be able to assess their progress to achieve optimum performance. Neurofeedback treatment protocols mainly focus on the alpha, beta, delta, theta, and gamma waves or their combination (5).

Landscape and Use Cases

Most start-ups developing BCI hardware tend to have a common focus, such as sleep, emotions and wellbeing, gaming, and productivity and focus. For example, Neurosity (est. 2018, US) has developed a non-invasive headset crown that measures brain waves, plays music that increases focus, and tracks performance. Non-invasive Muse (est. 2009, Canada) offers a wearable supporting meditation and sleep by providing real-time feedback on EEG brainwaves to optimize performance. Recently, Snap acquired French neurotech start-up NextMind to incorporate its non-invasive BCI into future AR glasses.

Several non-invasive start-ups also venture into more serious medical use cases. For example, Neuroelectrics attempts to treat brain disorders with personalized neuromodulation using their electric headcap, Starstim, which the team claims can significantly reduce seizures in patients with treatment-resistant epilepsy. They raised $17.5mm in 2021 and seek an FDA license in the US for their product.

On the invasive side, Neuralink (est. 2016, US) continues to make news, with the latest Elon Musk tweet on April 24 that Neuralink will “definitely” cure tinnitus which affects 50 million people in the US alone. The company provides the Neuralink app that would allow their users to control their iOS device, keyboard, and mouse directly with the activity of their brain, just by thinking about it. A bit more under the radar is Bios Health, which like Neuralink, has developed its implant but is more focused on the data collected from the device instead of making the device less clunky. The start-up wants to help treat diseases with no effective drugs by rewiring the brain.

Hardware Form Factor

All BCI systems depend on the sensors and associated hardware that acquire the brain signals. Although non-invasive solutions have come a long way from ultra-clunky hardware and what looked like a “swim cap with electrodes” to wireless wearables, many of the products available on the market are still compromised by signal quality and practicality.

Adoption by end-users primarily concerns ergonomics. More precisely, EEG-based (non-invasive) BCI solutions must “ideally {…} have electrodes that do not require skin abrasion or conductive gel (i.e., so-called dry electrodes); be small and fully portable; have comfortable, convenient, and cosmetically acceptable mountings; be easy to set up; function for many hours without maintenance; perform well in all environments; {…} interface easily with a wide range of applications” and be “inexpensive in terms of training time and dedicated resources”. As you can tell, the list of requirements for commercial-type devices is quite exhaustive. Even if a start-up can meet all these requirements, the required ongoing calibration for the device to provide accurate readouts is where most consumers lose interest.

However, the team at Neurable (est. 2015, US) seems to have developed a form factor that meets these criteria AND does not require any calibration. Their beachhead product “Enten” is an elegant headphone set that yields great signal acquisition and processing. The team has been working on BCI tech for a decade, evolving the product from a thought-controlled VR to the current non-invasive BCI headphone. Their soon-to-launch commercially available headphones provide data about your focus and work routines to help you make better decisions.

BCI Software

Beyond BCI hardware in the form of implants, headsets, or -bands, companies are leveraging neurotech software. One use case example is digital neurotherapeutics, using BCI in various forms using VR, AR, and motion capture game systems. See MindMaze (est. 2017) out of Switzerland, which raised money from Leo DiCaprio in 2017 and another $100m in February 2022. Arctop (est. 2016, US), whose founder, Dan Furman, worked with Stephen Hawking on developing his personal neural interface for communication, has developed NeuOS, an SDK (“Software Development Kit,” used by Developers) providing their patented technology that gives apps human-like intelligence by personalizing performance through various real-time data points (emotion, cognition, biometric data).

Ceregate (est. 2019, Germany) has developed a neural interface system that writes information directly into the brain. Their platform is hardware-agnostic and can be utilized to develop many therapeutic solutions. Their first software product is used to treat symptoms of Parkinson’s, and in January 2022, they raised a round from 468 Capital and re.Mind Capital (Apeiron Investment Group).

Ethics

Despite the fascinating use cases and developments listed above, BCI technology also raises ethical concerns. Particularly autonomy, safety, privacy, and security issues are at hand. Ethicists question, for example, whether an action that is produced primarily or solely by a device can genuinely be attributed to a human. A BCI could, for instance, take signal input directly from the brain and execute inappropriate actions that would usually be considered but not executed (6).

Moreover, in terms of privacy, a subject may be “unaware of the extent of the information obtained from his or her brain” (7). Beyond that, external sources gaining control of a BCI device whereby the BCI user could be exposed to inference from others may be a real threat (10). Such issues are significant for the future development of BCI technology.

Future Applications

We believe that BCIs will influence how we work in the long term. In the workplace, companies will want to leverage this tech to enhance productivity and improve decisions among operators and managers. EEG-based BCI could monitor operators’ mental states like fatigue, stress, or loss of vigilance which can be critical during dangerous activities. “In particular, fatigue monitoring is considered a valuable tool in repetitive and automatic tasks such as driving, piloting or quality control”.

Additionally, BCI could enable further breakthroughs for human health and longevity. Combining BCI tech with another biomarker-sensing tech is very exciting. For example, BCIs based on metabolic activity assessed by measuring blood oxygenation through functional magnetic resonance imaging (fMRI) or near-infrared spectroscopy (NIRS) correlate with neural activity (see our last article on spectroscopy). An fMRI-based BCI can thus enhance the real-time control of assistive technology such as robotic arms.

The BCI field is accelerating at an unprecedented pace, with easy-to-use devices entering the market that can measure brain activity and generate simple, real-time insights. This data will allow individuals to be more focused, avoid burnout, lessen mental stresses, sleep better, and be more conscious of themselves.

We believe that there has never been a better time to explore and advance BCI tech than now.

Sources

1. Shih, J.J., Krusienski, D.J., Wolpaw, J.R. 2012. Brain-Computer Interfaces in Medicine. Mayo Clinic Proceedings, 87(3).
2. Jabr, F. 2022. The Man Who Controls Computers With His Mind. The New York Times & Synchron Announces Enrollment of First Patient in US Endovascular BCI Study COMMAND in Patients with Severe Paralysis. BusinessWire
3. Rashid M., Sulaiman N. , et al. 2020. Current Status, Challenges, and Possible Solutions of EEG-Based Brain-Computer Interface: A Comprehensive Review. Front. Neurorobot, 14(25).
4. Wolpaw, J.R., McFarland, D.J. 2004. Control of a two-dimensional movement signal by a non-invasive brain-computer interface in humans. PNAS, 101(51).
5. Marzbani, H., Marateb, H.R., Mansourian, M. 2016. Neurofeedback: A Comprehensive Review on System Design, Methodology and Clinical Applications. Basic Clin Neurosci., 7(2).
6. Burwell, S. et al. 2017. Ethical aspects of brain-computer interfaces: a scoping review. BMC Medical Ethics, 18(60).
7. Vlek, R.J. et al. 2012. Ethical issues in brain-computer interface research, development, and dissemination. J Neurol Phys Ther, 36(2).