Unlocking the mind with mind-reading technology
By Hong Kong Economic Times
Have you ever imagined a world where, without us lifting a finger, our thoughts may turn into actions? In the film Iron Man 3, Mark 42, the armour of Tony Stark, the protagonist, responds to his mental command and acts out his thoughts. In March 2024, Neuralink, a brain-chip start-up founded by billionaire entrepreneur Elon Musk, livestreamed a video showing a man paralysed from the shoulders down moving a cursor on a computer to play chess after having a device implanted in him.
From ancient times to the present day, mind-reading has always been a mysterious object of fascin ation in human civilisations. Today, with the rapid advancement of technology, mind-reading, or the ability to access an individual’s mental states, is no longer a distant dream: it is gradually becoming a reality.
Much as Neuralink does, an increasing number of research institutions and companies are trying to commercialise this technology to help connect the brain directly with external devices, with applications not only in the medical sector but also in the military, education and the entertainment industry. This emerging mind-reading technology involves multiple disciplines, such as computer science, medicine and mechanics.
By integrating knowledge from artificial intelligence (AI) and big data to neuroscience, modern mind-reading technology can gain insights into an individual’s mental states, emotional fluctuations and even their intentions based on data analysis and algorithms.
However, the rise of such a technology also raised privacy and ethical concerns. A long distance is still ahead before we achieve the delicate balance between enjoying the convenience brought about by this technology and protecting personal privacy and individual rights.
Finding the balance between enjoying the convenience brought about by BCI technology and protecting personal privacy and individual rights is a key issue
Brain-computer Interfaces (BCIs) and neurotechnology
The brain is the most complex organ of the human body. It produces our thoughts, feelings and actions, and consists of about 86 billion neurons that, among themselves, form 100 trillion connections. These neurons connect and communicate with one another by sending electrical and chemical signals.
Scientists have been working on the BCI, or brain-machine interface (BMI), a computer-based system that allows users to interact with the environment by capturing their brain signals, analysing them and converting them into action commands for an output device1.
It is common to study brain activity’s electrical signals, which are commonly obtained through electrodes on the scalp, the cortical surface, or in the cortex.
A BCI system is either “exogenous” or “endogenous” depending on the input signals. Exogenous BCIs use brain responses triggered by external stimuli like visual evoked potentials or auditory evoked potentials, whereas endogenous BCIs rely on the self-regulation of brain rhythms and potentials without external cues. Endogenous BCIs allow free-willed operation and cursor movement in a two-dimensional space. Exogenous BCIs may limit users to predefined choices.
Based on their method of processing input data, BCIs can be categorised into synchronous or asynchronous systems. Synchronous BCIs analyse brain signals within a preset time frame and ignore other signals, so users can only send commands during specific periods determined by the system. In contrast, asynchronous BCIs continuously analyse brain signals, thereby allowing users to act whenever they want. Therefore, asynchronous BCIs offer a more natural mode of human-machine interaction because it provides greater user autonomy. The challenge is that asynchronous BCIs are more complex and require more computational resources2.
The brain is the most complex organ of the human body, producing our thoughts, feelings and actions
What these technologies can do
Initially, the demand for BCIs comes from the healthcare industry: rehabilitation, prosthetics control, and neuro-feedback.
The functions of patients suffering from physical impairments can be restored by rehabilitative BCIs to a certain limited extent. Through monitoring Electroencephalography (EEG) signals, electrophysiological abnormalities for patients with neurodevelopmental disorders can be identified, and the emotional status of patients with mental disorders can be analysed.
BCIs can be an effective solution to epilepsy, amyotrophic lateral sclerosis (ALS) and Parkinson’s disease. Take epilepsy as an example: It is one of the commonest neurological diseases, affecting around 50 million people globally. The disease is characterised by recurrent seizures, the result of excessive electrical discharges in a group of brain cells.
For drug-resistant patients, surgical solutions, such as the implantation of brain pacemakers, can effectively control epilepsy symptoms. Medical tech company Medtronic’s PerceptTM PC Deep Brain Stimulation system, the first commercially available implantable pulse generator that can store the day-to-day brain activity of patients with implanted electrodes, can chronically capture and record brain signals with BrainSense technology while delivering therapy to patients with neurological disorders, allowing physicians to track the brain signals and correlate them with patient’s actions or experiences3.
Rehabilitative BCIs are used in the medical field, helping patients restore their functions to a certain extent
Helping patients suffering from paralysis or motor impairments regain motor abilities has always been one of the most important pursuits of BCI. The research focus is on the restoration of patient’s motor functions such as controlling robotic arms to grasp objects or moving a cursor and selecting letters to input on a computer.
Over the years, progress has been made on this front but difficulties remain. The restoration of hand movements is especially challenging because its greater complexity and flexibility (compared with leg movements) make it more difficult to be controlled with BCI.
In 2020, researchers at Battelle Memorial Institute and the Ohio State University Wexner Medical Center, using a BCI system, helped restore sensation to the hand of a research participant with severe spinal cord injury. Although the participant had had almost no sensation remaining in his hand, there was a small neural signal when his skin was stimulated, indicating that even in those who have suffered from a “clinically complete” spinal cord injury, there are almost always a few wisps of nerve fibres that remain intact. This allows researchers to boost these signals to such a level that the brain can receive and respond to them4.
In the consumer domain, non-invasive brain-machine interface hardware devices have begun to assimilate into our daily lives. Various wearable devices now help users de-stress, improve focus and alleviate sleep disorders. These devices are suitable for individual homes and hotels, since they can monitor the user’s sleep status by monitoring their EEG signals and even control smart home devices accordingly.
Private companies are also working to unlock the potential of the human brain with BCIs. By reading the EEG signals in real time, BCI systems can provide timely visual and auditory feedback, assisting professional athletes with their training in concentration and performance enhancement.
With its development, BCI technology will continue to enhance the quality of life of patients with movement disorders, and is expected to be applied in more fields in the future.
History and development
The development of BCI is intimately linked to the advancement of neuroscience. Back in the 19th century, British physician Richard Caton first discovered the waves of electrical potential by observing the electrical impulses from the brain surfaces of living animals. His discovery formed the basis of EEG.
In the early 20th century, the concept of BCI, or rather, what we today understand retrospectively and perhaps anachronistically as BCI, evolved significantly as pioneering research in neurophysiology and computer technology laid the groundwork for the BCI we are familiar with today.
With the 1920s came the invention of electroencephalogram by German psychiatrist Hans Berger. It was a historical breakthrough that allowed researchers to record the electrical activity of the human brain. Since then, there have been speculations that it can be used for communication and control.
In 1969, researcher Eberhard Fetz at the University of Washington first showed that a monkey can control the needle of a metre with its thoughts alone5.
In the 1973, research on BCIs can be said to have really begun at the University of California, Los Angeles (UCLA), as computer scientist Jacques Vidal coined the term “BCI” in a published paper titled “Toward Direct Brain-Computer Communications”6.
Initially, BCIs experiments were conducted on primates. In 1998, US neurologist Philip Kennedy implanted the first invasive BCI into the brain of a man who was paralysed due to a brain stem injury.
The first International BCI Meeting was held in 1999, with 50 scientists from 22 laboratories attending.
In the 21st century, brain-machine interfaces have entered the clinical application stage. In 2016, a team of researchers from the University of Pittsburgh, working with the University of Chicago, designed the brain implant that allows a patient to sense touch via a robotic hand7.
As of December 2019, more than 760,000 people worldwide have cochlear implants as the neuroprosthetic device assisting with hearing restoration8.
The BCI solution by Elon Musk’s Neuralink is among those with better commercialisation prospects. The company has completed its first human implantation in January 2024 after gaining approval from the United States Food and Drug Administration in May 20239.
There are also other companies developing BCIs. Backed by funding from Microsoft co-founder Bill Gates and Amazon.com founder Jeff Bezos, New York-based Synchron has implanted its device in ten individuals. Austin-based startup Paradromics has developed chips designed to penetrate the cortex10.
Elon Musk, the founder of the startup Neuralink, announcing that the first human patient implanted with a brain-chip can control a computer mouse using their thoughts
Neurotechnologies including BCIs have become a subset of the frontier technologies that countries are focusing on, with medical use being the most important application scenario. Countries like the United States and China are pioneers in the research and development of BCIs. The former launched the multibillion-dollar Brain Research Through Advancing Innovative Neurotechnologies (BRAIN) initiative in 2013 to accelerate the development of innovative neurotechnologies.
China also launched its brain initiative by listing brain science and brain-like research as a national major scientific and technological innovation and engineering project in its 13th Five-Year Plan in 2016.
Despite a late start, China filed about 35% of the global patents in brain-computer technology in recent years, outpacing the United States (30%) and Japan (10%), according to a report by China’s Brain-Computer Interface Industrial Alliance in May 202311.
Countries including Japan, South Korea and Canada also followed suit to roll out similar initiatives.
Researchers in Hong Kong have made significant strides in the field of BCIs. In 2010, The Chinese University of Hong Kong showcased the first BCI system for Chinese character input at the China Hi-Tech Fair in Shenzhen12.
In 2018, biomedical engineers from The University of Hong Kong developed technologies that enable patients to control computer activity with their thoughts. With a cap of electrodes that transmitted their brain signals to the computer, the patients were able to select characters from a keyboard and move a mouse around the screen13.
The Hong Kong University of Science and Technology has set up a Computational Cognitive Engineering Lab to spearhead the revolution of the BCIs.
Accuracy and limitations
The accuracy of BCIs varies with the way the signals are collected and the decoding algorithms. In general, non-invasive BCIs, like EEG electrodes placed on the scalp, are less accurate than implanted BCIs.
Thanks to the help of AI, BCIs can now read our mind with surprising accuracy. A recent study by a team of researchers from Kobe University in Japan showed a remarkable 95% accuracy rate in predicting mouse movements based on brain functional imaging data. They used an AI image recognition algorithm that combines spatial and temporal pattern recognition algorithms14.
The limitations and risks of invasive BCIs are also obvious because the procedure involves implanting electrodes to collect and decode electroencephalographic signals. To insert the device into the brain tissue, doctors would need to create a hole in the skull. Apart from the technical difficulty, invasive BCIs pose a risk of infection. In the case of cranial infections, electrode failure or end of electrode lifespan, the electrodes must be removed from the brain to avoid secondary damage.
Moreover, as time goes by, electrodes of the implanted device may be covered by connective tissue, which could lead to a gradual weakening or even disappearance of signals.
Ethical issues
Emerging brain-reading technologies like BCIs can no doubt bring exciting possibilities and unprecedented changes, since people will be able to expand the brain with an external device while uploading their thoughts in real time.
But since BCIs connect the human brain with machines, which can potentially affect brain function, profound ethical concerns have been raised. These issues touch upon safety, autonomy and privacy, presenting moral and legal challenges. A comprehensive ethical framework is required to protect individual rights with the rise of concerns regarding privacy, cognitive liberty and equity, and self-conception.
In 2019, a trial of a “mind-reading” gadget was halted in a Chinese primary school after online backlash against it. The head-mounted device called FocusEDU uses EEG sensors to detect brain activity and share relevant data with parents. It aims to monitor students’ engagement in class and train their concentration, but netizens have raised doubts over the product’s safety and the infringement of privacy15.
Earlier in 2024, the NRC adopted an ethical guideline for BCI research published by the National Science and Technology Ethics Committee. It is the first guideline of its kind in the country.
The guideline specifies that research on BCI should cause no damage, and its fundamental purpose should be assisting, enhancing and repairing sensory-motor functions or improving human-computer interactions for human health and well-being16.
In the US, the Food and Drug Administration issued, in 2019, a draft non-binding guidance to provide recommendations for submissions regarding BCI devices for patients suffering from paralysis or those who have undergone amputation17.
In September 2023, South Korea submitted a proposal on the global standardisation of interface data between the human brain and computers to the joint technical committee of the International Organization for Standardization (ISO) and the International Electrotechnical Commission on BCI18.
The role of engineers
Engineers play a pivotal role in the development and advancement of mind-reading technology because neural engineering is one of the most essential technologies and requires inter-disciplinary knowledge of neuroscience, computation and robotics.
The global BCI market is expected to be worth US$8.9 billion by 2032, according to a report by the Acumen Research and Consulting in India. The Asia-Pacific region is projected to witness substantial growth, with a Compound Annual Growth Rate (CAGR) of over 17% from 2023 to 203219.
Although medically used BCIs will remain the focus for its ability to address the needs of patients with neurological disorders, BCIs application is gradually expanding from the medical field to industries and entertainment. This indicates tremendous opportunities for the engineering industry, especially those who work in neuroscience, machine learning and signal processing.
To promote the commercialisation of BCIs, engineers can work on addressing the challenges in BCIs, such as low data transfer rates, high cognitive load, high error rates in signal processing, and security and ethics.
Engineers can make contributions by designing and developing hardware and software, improving algorithms to process and translate brain signals, and ensuring devices’ compliance with medical standards.
Engineers will also need to work closely with neuroscientists, doctors and law experts to bridge the gap between theory and practice, between concepts and applications.
Neuroscience is one of the most sophisticated and challenging research fields. The essence of mind-reading technology is brain-reading. With the rapid development of biotechnology and information technology, BCIs can help reflect ideas in the human brain to the digital world and even realise them. A technological revolution is underway and the expertise of engineers is shaping the future of BCIs.
References
- Shih J J, Krusienski D J and Wolpaw J R (2012). ‘Brain-computer interfaces in medicine’. Mayo Clinic proceedings, 87(3), pp. 268–279. doi: https://doi.org/10.1016/j.mayocp.2011.12.008. (Accessed: 29 April 2024)
- Nicolas-Alonso L F and Gomez-Gil J (2012). ‘Brain Computer Interfaces, a review’. Sensors , 12(2), pp. 1211-1279. doi: https://doi.org/10.3390/s120201211 (Accessed: 29 April 2024).
- Medtronic Canada Newsroom (2020). PerceptTM PC Deep brain stimulation with BrainSenseTM Technology. Available at: https://canadanews.medtronic.com/PerceptTM-PC-Deep-Brain- Stimulation-with-BrainSenseTM-Technology (Accessed: 29 April 2024)
- ScienceDaily (2020). Researchers restore injured man’s sense of touch using brain-computer interface technology . Available at: https://www.sciencedaily.com/releases/2020/04/200423130508.htm (Accessed: 29 April 2024)
- The Economist (2018). Eberhard Fetz . Available at: https://www.economist.com/technology-quarterly/2018/01/04/ eberhard-fetz (Accessed: 29 April 2024)
- UCLA Samueli Newsroom (2019). Cerebral connections: UCLA engineers tap into rich legacy of brain-computer interface technology, UCLA Samueli School Of Engineering . Available at: https://samueli.ucla.edu/brain-computer-interface/ (Accessed: 29 April 2024)
- Thometz K (2016) Paralyzed man regains sense of touch with robotic arm . WTTW News. Available at: https://news.wttw.com/2016/10/18/paralyzed-man-regains-sense-touchrobotic- arm (Accessed: 29 April 2024)
- National Institute of Deafness and Other Communication Disorders. Cochlear implants. Available at: https://www.nidcd.nih.gov/health/cochlear-implants#:~:text=over the telephone.-,Who gets cochlear implants?,adults and 65,000 in children (Accessed: 29 April 2024)
- Oi M (2024). Neuralink: Musk’s firm says First Brain-chip patient plays online chess . BBC News. Available at: https://www.bbc.com/news/business-68622781 (Accessed: 29 April 2024)
- Willyard C (2024). Beyond neuralink: Meet the other companies developing brain-computer interfaces. MIT Technology Review. Available at: https://www.technologyreview.com/2024/04/19/1091505/companiesbrain- computer-interfaces/ (Accessed: 29 April 2024)
- Cheng Y (2023) Brain-Computer Tech on March in country. Chinadaily.com.cn. Available at: https://www.chinadaily.com.cn/a/202305/31/ WS64767b89a3107584c3ac2fe0.html (Accessed: 29 April 2024)
- The Chinese University of Hong Kong Communications and public relations office (2010). CUHK showcases first ever brain-computer interface for Chinese character input at Hi-Tech Fair . Available at: https://www.cpr.cuhk.edu.hk/en/press/cuhk-showcases-first-everbrain- computer-interface-for-chinese-character-input-at-hi-tech-fair/ (Accessed: 29 April 2024)
- The University of Hong Kong (2018). UP FOR GRABS. HKU Bulletin. Available at: https://www4.hku.hk/pubunit/Bulletin/2018_May_Vol.19_No.3/cover_ story/page5.html (Accessed: 29 April 2024)
- McMillan T (2024). Breakthrough in brain-computer interfaces: Scientists use AI neural decoding to predict mouse movements with 95% accuracy. The Debrief. Available at: https://thedebrief.org/breakthrough-in-brain-computer-interfacesscientists- use-ai-neural-decoding-to-predict-mouse-movements-with-95- accuracy/ (Accessed: 29 April 2024)
- Standaert M (2019). Chinese primary school halts trial of device that monitors pupils’ Brainwaves . The Guardian. Available at: https://www.theguardian.com/world/2019/nov/01/chinese-primaryschool- halts-trial-of-device-that-monitors-pupils-brainwaves (Accessed: 29 April 2024)
- Xinhua (ed.) (2024). China adopts ethical guideline for Brain-Computer Interface Research. Xinhua News Agency. Available at: https://english.news.cn/20240206/20ee5b3b5093413cae5d2c617be7 4d10/c.html (Accessed: 29 April 2024)
- The United States Food and Drug Administration. Implanted Brain-Computer Interface (BCI) Devices for Patients with Paralysis or Amputation - Non-clinical Testing and Clinical Considerations . Available at: https://www.fda.gov/media/120362/download (Accessed: 29 April 2024)
- Ko D (2023). Korea to set global standards for Human Brain-Computer Interface Data . Korea Times. Available at: https://www.koreatimes.co.kr/www/tech/2024/04/129_358905.html (Accessed: 29 April 2024)
- Acumen Research and Consulting (2023) Brain Computer Interface Market is forecasted to reach USD 8.9 billion by 2032, growing at a 16.8% CAGR from 2023 to 2032. GlobeNewswire News Room. Available at: https://www.globenewswire.com/news-release/2023/10/05/2755594/0/ en/Brain-Computer-Interface-Market-is-forecasted-to-reach-USD- 8-9-Billion-by-2032-growing-at-a-16-8-CAGR-from-2023-to-2032.html (Accessed: 29 April 2024)