Starter Toolkit Series: Brain Computer Interfaces (BCI)
Part of my series demystifying basic concepts of new-age technologies to help spark curiosity in other young minds
What are BCIs?
Brain-Computer Interface (BCI): devices that enable its users to interact with computers by means of brain-activity only, this activity being generally measured by Electroencephalography (EEG)* or fMRI* based technologies but has future potential for non-invasive non-contact measurement.
*EEG — allows recording of brain activity using electrodes placed on the scalp to target specific areas of interest in the brain.
*fMRI — measures blood flow in the brain
How do they work?
Before I start going on and on about BCIs, you should first understand the basics of the brain.
The brain is made up of various cortexes. These cortexes control different parts of how we live. Some of these are:
- Visual cortex — processes visual signals
- Auditory cortex — processes sound
- Somatosensory cortex — processes touch
- Motor cortex — plans and executes physical movements
The human brain also contains approximately 100 billion neurons. The neurons use positively and negatively charged ions to communicate with each other through connectors called axons and dendrites. Such communication is interpreted by the human body in an extremely complex, and not completely understood, series of reactions that affect actions and other reactions in the rest of the body. These signals are how humans perceive senses of touch, sight, pain, etc.
Each neuron transitions between stages of ‘rest’ and that of ‘activation’. During rest negatively charged ions are inside the neuron’s membrane with more positively charged ions outside the membrane. The neuron at this time has a negative charge. When activated by the brain, the positively charged ions enter the neuron via the membrane, replacing the negatively charged ions that are sent out of the neuron, in turn triggering neighboring neurons. This electrochemical activity in the brain’s neurons is the fundamental basis of the brain’s electrical activity.
The signals that are currently leveraged for BCI purposes are Spike signals and Field Potentials, both kinds of signals are detectable and processable by algorithms to allow interpretation. Signals are detectable using EEG or fMRI, both methods offer varying clarity of speed and accuracy in signal detection.
Signal detection using either method includes a lot of ‘noise’, which are electrical signals caused by other activities in the human body potentially not originating in the brain. Advances are being made in the use of artificial intelligence and Machine Learning to interpret, cleanse and use the signals.
How are they used now?
BCIs today are mainly used for a technology called Trans-Cranial Direct Current Stimulation (tDCS). tDCS uses a non-invasive brain stimulation that seeks to influence parts of the brain’s electrical activity .
During a typical tDCS treatment, patients are conscious and seated while small sponge electrodes are placed over specific parts of their brain. A small electric current is then passed through, via the electrodes, that gently stimulates the brain. This electrical current affects a targeted part of the brain, alters the electrical activity of all the neurons in that targeted brain area, which could for instance, be the Motor Cortex thereby potentially modifying the electrical signals produced by the neurons in that area of the brain. After treatment, patients perform specific exercises that can then retrain the part of the recently stimulated brain. Deeper learning of interpretation of signals from the brain has shown tDCS to help in reducing feelings of Depression, even as treatments for more serious Brain injuries, Chronic pain, Dementia, Fibromyalgia, Migraines and Strokes.
Special advanced EEG scans allow targeted interpretation of signals from specific areas of the brain that further informs targeted treatment via tDCS.
Mitigating effects of Depression
Depression is considered to be caused by imbalance in brain activity in the left (lower activity) and right (excessive activity) ‘dorsolateral prefrontal cortex’ (DLPFC). The DLPFC has connections to deeper areas of the brain that affect mood and stimulation of the left DLPFC is shown to have improvements in Depression symptoms.
ADHD — Attention Deficit Hyperactivity Disorder
ADHD is a disorder that causes difficulty in concentration, inability to ignore distractions and lack of impulse control. It is considered that reduced electrical activity in the DLPFC is a causal factor in triggering ADHD symptoms and tDCS based treatment has shown improvement in subjects with ADHD.
Cognitive learning abilities
Applying tDCS to the right parietal lobe, the right anterior temporal lobe (rATL), the left anterior temporal lobe (lATL) and the frontal lobes of the brain has been shown to improve cognitive learning ability of patients, increased problem solving abilities and improved memory respectively.
Potential Uses in the Future
Imagine having the ability to constantly access what your brain is thinking, factually without it being colored by your own perception, and potentially to people other than yourself. Now imagine you are able to do that without wires and electrodes sticking out of your head doing this measurement with EEG, techniques that are remotely controlled, not invasive, sensors that can analyze the signals with apps on your phone and deliver information to you, useful information, from the million data points being fed to it constantly.
We have already accepted a Fitbit or an Apple Watch measuring our heart rate and activity levels constantly, constantly providing us actionable information, ‘time to move’, ‘time to walk’ etc. The future holds the key for your phone or your watch telling you, ‘Time to Focus, Arjun’, ‘Your concentration level in the past 15 mins has been 56% of max capacity’, ‘Your brain signal shows increased stress level, Time to take a break’, etc
The next level of BCI will also include a feedback loop that will intervene and effect changes in our brain activity. Think of an app on your phone to analyze brain activity and reiterate the focus back when your mind loses focus, essentially ‘super-charge’ your brain for 30 mins — would come in real handy the night before the final term paper is due, wouldn’t it?
Another area of significant potential is in the way humans interact with machines. We’re looking at a future where a keyboard or a mouse is made redundant, if computers or other devices can receive and interpret signals directly from the brain.
One of my personal heroes, Elon Musk, has driven a lot of prime-time attention to the potential of exploiting BCI technologies for human benefit with his new company Neuralink (link attatched). Neuralink is developing ‘The Link’ an implanted device that can record the activity of thousands of neurons in the brain, interpret and send the signals to the Link, that can then use the signals to drive real world devices such as a mouse, robotic devices that can move the leg of , for instance, a paraplegic using signals read from her brain read via the Link. You can visit Neuralink’s website for more information and to remain updated on the latest developments.
BCI-based devices will make a world of difference in the quality of life of the differently-abled. There’s obviously a long way to go, research is still in early stages even on how much readable and interpretable signals humans are even able to produce. But where the tech is taking us will make BCI-based devices ubiquitous and life-saving in the near future.
Ethical Concerns
The more we leverage the ability to read, interpret and influence brain signals, the more we open ourselves up to the possibility of mis-use of such powerful technology.
Neuroethics is a branch of the overall ethical spectrum of evolving technologies that particularly focuses attention on technologies that impact the brain. There is significant research, dialogue and collaboration today with existing deployments of BCI in the field of medicine including patients with epilepsy or comatose being monitored via EEG devices to alert medical professionals about potential emergency situations that can then trigger medical corrective actions automatically,
With potential future developments that increasingly leverage brain signals non-invasively, there is the ever persistent threat of hacking of brain signals, and worse, hacking of the feedback loop giving malicious players access to influence behavior or response of an individual’s brain activity. Additionally other privacy concerns relating to development of BCI-based technologies are an area of increasing global attention.
What Next?
Interested in applying your own AI based algorithms to the analysis of brain signals? You can find a public database of brain signals here
Some good further reading can be found at the links below:
