What is a brain-computer interface? An article to help you understand!

Brain-computer interface applications, which refer to devices that allow users to interact with computers to measure users’ brain activity, have accelerated the transformation of the medical device industry with their unique advantages and will continue to do so. Electroencephalography (EEG), which identifies energy and frequency patterns in the brain, has become a common method for measuring electrical activity in the scalp. Artificial intelligence (AI) and machine learning (ML) can improve the accuracy and reliability of evaluating and developing brain-computer interface applications. Next, we’ll dive into this area and introduce some of the key Electronic components in the signal chain required to measure brain waves, and how AI integrates these components.

Author: Paul Golata

Writing is a delicate thing, and when you run out of inspiration and can’t write a word, you may even feel like the blinking mouse is laughing at you. To pass the time, I started recording the number of mouse blinks in a minute and concluded that it blinked at 50Hz. Every flicker of it seems to tell me that I’m still doing nothing so far. My brain went blank, like someone wiped my memory with an amnestic from the movie Men in Black (1997). For most writers, it’s often frustrating how to start a good story, how to write an essay that engages readers and delivers a message. How can I express my thoughts in words? This requires a collision of ideas again and again.

If only my brain could be intelligently augmented with a wealth of valuable data with the aid of a brain-computer interface (BCI), sparking an instant stream of consciousness on a computer screen via a keyboard.

Of course, I’m not trying to discuss my writing with you. Brain-computer interface is the topic of this article, so let’s get back to the point.

The concept of brain-computer interface

Brain-computer interface applications, which refer to devices that allow users to interact with computers to measure users’ brain activity, have accelerated the transformation of the medical device industry with their unique advantages and will continue to do so. Electroencephalography (EEG), which identifies energy and frequency patterns in the brain, has become a common method for measuring electrical activity in the scalp. Artificial intelligence (AI) and machine learning (ML) can improve the accuracy and reliability of evaluating and developing brain-computer interface applications. Next, we’ll dive into this area and introduce some of the key electronic components in the signal chain required to measure brain waves, and how AI integrates these components.

Brain-computer interface and EEG

The human brain produces oscillating voltages, typically very small, measured in millionths of a volt. For this reason, these brainwave voltages are typically collected and analyzed using EEG, an electrophysiological monitoring method that records electrical activity in the scalp to capture signals directly related to the brainwaves that occur just below the skull (Figure 1).

What is a brain-computer interface? An article to help you understand!
Figure 1: A woman undergoing a noninvasive EEG (Credit: Yakobchuk Viacheslav/Shutterstock.com)

Brain-computer interface communication via EEG can be one-way or two-way. Two-way communication allows information to flow in both directions, providing feedback to the brain and making further adjustments. EEG can use non-invasive, semi-invasive and invasive techniques: invasive EEG refers to implanting devices directly into the brain and making connections; semi-invasive EEG refers to placing electrodes on the surface of the cerebral cortex; and Non-invasive is usually accomplished by placing a cap with various electrodes on the skull. The EEG has a temporal resolution of 0.05 seconds and a spatial resolution of around 10 mm. In addition to the electrical techniques used in EEG, other techniques such as magnetism and metabolism can also be used to collect data.

Brain waves are generally classified into five categories by frequency (Table 1), and medical researchers divide them into five bands, each of which corresponds to a different brain state. For example, critical activities such as memory and recall are usually processed in the theta band, but there are exceptions. Researchers use these bands to analyze what can happen when a signal is too small, too large, or at optimal values.

Table 1: Brain waves are divided into five categories, and each frequency range is a nominal value, not an absolute value. (Source: Author)
What is a brain-computer interface? An article to help you understand!

An EEG captures and digitizes brain wave signals, then performs signal processing on them, extracting features and classifying them using translation algorithms. Of course, it can also be printed or recorded for future analysis. Signal outputs can be used to generate device commands to provide instructions related to motor control, motion/movement, and environmental conditions or stimuli. Therefore, brain-computer interfaces can help disabled people to better control the external environment.

Brain-Computer Interfaces and the Human Condition

Our senses and intelligence are fundamentally limited due to human biology. It is conceivable that brain-computer interfaces and implants could augment or provide new sensory information, enhancing biological capabilities. Humans and machines operate in different ways, so special attention needs to be paid to how to achieve real-time extraction and exchange of information when merging. Beyond that, much else is needed to understand how the mutual coordination between the individual addressable neurons that guide the brain functions in the environment, while maintaining coordination with the interface’s digitally addressable domains.

Brain-computer interface research

The current research direction of brain-computer interface is to help humans, including restoration or enhancement of human vision, motor recovery of disabled limbs, and brain positioning to help restore and correct various neurological injuries and diseases. Brain mapping can help us better understand how human thinking translates into human behavior, which in turn can inspire enhanced learning, new or enhanced human perception, and new embedded autonomic nervous systems. Let’s look at two examples of where brain-computer interfaces can continue to evolve and dominate.

superhuman cognition

Billionaire Elon Musk is actively working on brain-computer interfaces. He is one of the founders of Neuralink, a neurotechnology company developing implantable brain-computer interfaces, dedicated to creating advanced brain-computer interface tools. The founders of the company believe that with the right team, the application of this technology will be even more promising. Neuralink is exploring the feasibility of implanting ultra-thin electronic wires that allow neural activity into the brain.

One of Musk’s stated goals is to achieve “superhuman cognition.” Inventor and futurist Ray Kurzweil believes that artificial intelligence is characterized by extraordinary pattern recognition abilities. He sees artificial intelligence as an evolved self-organizing hierarchical system operating in the context of biological pattern recognition machines. Musk’s aim is to achieve superhuman cognition, as he wants humans to be more adept at negotiating to understand the emergence and spread of more powerful artificial intelligence, which is getting better at pattern recognition.

new perception

When looking at how brain-computer interfaces can enhance human perception, consider the case of Neil Harbisson, founder of the Internet Foundation. He is recognized as the world’s first cyborg. Haverson, who was born colorblind, had an antenna permanently implanted in his skull to distinguish colors by listening, compensating for the visual limitations with hearing. For this reason, Harvison actively advocates the concept of future life that integrates technology into the human body.

sensor hub

To support the development of brain-computer interface technology, humans have adopted sensor hubs to collect other biometric information beyond brain-computer interfaces in various ways. The sensor hub uses multiple sensors and a microcontroller to collect and analyze numerous body parameters that are not directly accessible by brain waves, including gathering information about the body’s pulse, heart rate, pulse oximetry (SpO2), and estimated blood pressure.

high quality data

Because the brain signals are very weak, the entire electronic signal chain needs to be designed with priority to reduce noise, spurious and spurious signals. Patients may induce these problems through body movement, sweating, eye movements, heart rhythm, and more. 50Hz/60Hz noise, electrode skin contact issues, and cable movement can all contribute to electrical errors.

Considering the above reasons, we should try our best to choose signal chain products with high precision, low noise and high resolution when selecting electronic components. Therefore, low-noise amplifiers (LNAs), unity-gain buffers, and precision analog-to-digital converters (ADCs) are selected to prevent the introduction of unwanted signals while providing the ability to interpret the data accurately and reliably. The use of differential amplifiers and band-pass filters can also ensure high-quality data transmission.

ADC Recommendations for EEG

Here we recommend an ADC for EEG signal chain designers, Analog Device’s AD7177-2 32-bit sigma-delta analog-to-digital converter (ADC) (Figure 2). The devices in this family are low noise, fast settling, multiplexed 2-/4-channel (full/pseudo-differential) ADCs for low bandwidth inputs. The AD7177-2 has a maximum channel scan rate of 10kSPS (100µs) for fully settled data. Its output data rates range from 5SPS to 10kSPS. The AD7177-2 ADC integrates key analog and digital signal conditioning blocks, allowing designers to individually configure each analog input channel used.

What is a brain-computer interface? An article to help you understand!
Figure 2: Analog Devices Inc. AD7177-2 32-bit Σ-Δ ADC (Image source: Mouser Electronics)

In addition, there are Texas Instruments’ ADS1299-x 24-bit analog-to-digital converters (ADCs), a family of 4/6/8-channel, low-noise, 24-bit simultaneous sampling delta-sigma analog-to-digital converters (ADCs) (Figure 3 ). The ADS1299-x integrates all common functions for extracranial EEG and electrocardiography (ECG) applications with a high level of integration and high performance, enabling the creation of scalable medical device systems with significantly reduced size, power consumption, and overall cost .

What is a brain-computer interface? An article to help you understand!
Figure 3: Texas Instruments ADS1299-x 24-bit analog-to-digital converters (Image credit: Mouser Electronics)

Artificial intelligence and brain-computer interface technology

Subfields such as artificial intelligence and machine learning and deep learning (DL) support EEG-based brain-computer interfaces (Figure 4). Deep learning provides tools for automatically classifying EEG signals, using the data for various applications and other convolutional neural network (CNN) training. Humans’ existing expertise is sufficient to support AI technology. Our desire is to remove artifacts, improve data quality, and continue to achieve advances in AI technology that can be classified by AI through DL techniques as measured brainwave signals continue to grow exponentially.

What is a brain-computer interface? An article to help you understand!
Figure 4: The evolution of artificial intelligence, machine learning, and deep learning (Source: elenabsl/Shutterstock.com)

in conclusion

Brain-wave-based brain-computer interface technology relies on high-performance electronic signal chains. Careful selection of all critical electronic components required to measure brain waves and other bodily functions is critical to improving design reliability. AI and DL technologies not only facilitate better interpretation of dynamic brainwave data, but also help humans gain more benefits from brain-computer interfaces. Brain-computer interface is an emerging method of human-computer interface, and will eventually provide a treatment for the psychological obstacles we encounter, especially the problem of creative exhaustion that writers face.

The Links:   MSP430F5328IRGCT NL12876BC26-21