The Basics of Brain Computer Interfaces
Many of you have seen videos on the internet of someone moving an RC car or seeing an amputee lift their robotic arm just by thinking about it. The human brain can suddenly connect with these devices. How does this work? What technologies play a role in these mechanisms?
Brain-Computer Interfaces
Brain-Computer Interfaces (BCIs) are systems that connect the human brain to external technologies, giving users the ability to interact with the mechanisms — just by using brain-activity.
BCI systems acquire brain signals, analyze them, and translate them into commands that are performed by output devices, carrying out a requested action by the brain. Brain-activity is usually measured by an Electroencephalography (EEG), then further classified based on the users needs — these are topics we’ll be looking into a little later.
Working with Brain Signals
To start off, we can say that the brain is divided into 2 portions: The Limbic System, and The Neocortex.
The Limbic System is a complex brain system responsible for managing psychological responses to emotional stimuli, eating and reproducing. In other words, this system has to do with everything related to survival and our primal urges, along with our attention, behaviour and character.
The Neocortex is responsible for all of the more complicated tasks. Some of the things the Neocortex takes care of are perception, reasoning, language, commands and decision making. In other words, it has to do with everything related to the logical functions in our everyday lives.
In the brain, we have approximately 86 million neurons (nerve cells), all individually linked to other neurons (through axons, the connectors). Whenever we move, think, or feel, neurons are working. For small tasks or large tasks, small electric signals move from neuron to neuron, doing all of the work. When we use BCIs, spiking patterns in signals are primarily used to give us information about what’s going on inside the brain.
What do I mean by spike patterns? If you’ve seen brain signals or heart monitors, you might’ve noticed the “spikes” on the screen (zig-zag lines.) Spike patterns are used to analyze brain data sets.
But how are we supposed to measure these spike patterns?
Some of the main tools that are used to measure brain signals include Electroencephalography (EEG) and Functional Magnetic Resonance Imaging (fMRI).
EEG is the physiological method used to record electrical activity generated by the brain. Electrodes are placed on the scalp surface to gather brain signals. EEG is the most common tool to use when working with BCIs. fMRI detects brain activity by analyzing changes in the blood flow of the brain.
Electrode: “A conductor through which electricity enters or leaves an object, subject, or region,” in this case, the brain.
So, I just use an EEG to detect spikes and there’s my BCI? It’s that simple?
When working with brain data, we tend to get a lot of extra information we don’t need. In fact, one of the biggest problems when working with EEG is that it can get very hard to filter out all of the unnecessary noise, such as signals from eye movement or teeth grinding, when really, you’re trying to capture the signals produced from hand movement.
EEG tends to show a broad range of brain signals, but in order to filter these signals, experts are using Machine Learning (ML) and Artificial Intelligence (AI). ML and AI help out by creating algorithms that run on data sets, helping in the filtering process. Additionally, signal conditioning is required to filter out environmental noises. For example, in USA, 60 Hz noise from electrical lines is very common, so working with brain signals can get quite complicated.
Types of BCIs
There are different types of BCIs, some used for simple tasks, others for more complicated tasks. The following techniques are used to capture brain signals:
Non-Invasive: These techniques are usually the safest and cheapest, but these devices only capture the “weaker” brain signals. With non-invasive devices, brain signals are detected by placing electrodes on the scalp.
Semi-Invasive: In these techniques, the devices are inserted into the skull (on top of the brain).
Invasive: In these techniques, special devices have to be inserted directly into the brain through a complex surgery. Invasive techniques are usually the most risky and expensive, but they work the best and can perform complex tasks.
Capabilities and Challenges
Now that we’re done with the understanding portion, what can BCIs actually do? What challenges are scientists facing when working with BCIs?
Well, BCIs are still in the research and development stages. As of today, they are not meant for humans in our day-to-day lives. However, once developed, BCIs can make living better. BCIs are capable of replacing/restoring useful body parts and functions for people severely disabled (e.g. walking and speaking), improve rehabilitation for people with strokes, restore visual perception to the blind, and more. We’ve already seen videos of robotic arms being controlled by the brain, just imagine how much more we can do.
Some challenges that come with BCIs are the inaccessibility, inaccuracy, and ethical issues. Firstly, BCI systems aren’t the easiest to carry around. All complex BCI systems in development are located in sophisticated environments, not available all the time. Also, EEG headsets aren’t the most comfortable either — having electrodes stuck on your head while performing everyday tasks can be quite irritating. Secondly, BCI systems still require a lot of development to be used in our everyday lives. They aren’t the most accurate for many complicated tasks and many studies have to be done on the person who plans on using the system. Finally, some may not want a computer to analyze their thoughts and may be concerned about the privacy issues and question where their brain data is going. Some may not trust the system as it is connected to their brain, the most important part of the body.
Overall, Brain-Computer Interface systems require a lot of development but once developed, they can change millions of lives around the world.
Thanks for reading! If you’d like to subscribe to my monthly newsletter, click here, and if you’d like to connect with me on LinkedIn, click here!
