Brain Computer Interface Technology

By: A. Sethi, K. Yadav, Konstantinos Psannis

The working of the human brain has often intrigued neuroscientists and researchers and inspired them to replicate its work to develop modern concepts such as neural networks. One of the most recent and promising developments in this category is Brain Computer Interface or BCIs [1]. A Brain Computer Interface is a setup that collects signals, analyzes them, and then translates them into commands and actions that are communicated to a device that displays the desired output. This mode of conveying a message is unique as it deviates from the brain’s conventional routine of transferring the output signal by using the course of peripheral nerves and muscles. This development can be considered as a miracle for those who suffer from neuro-muscular impairments.

BCIs can be broadly classified into two distinct categories based on the electrodes used to calibrate brain activity – invasive and non-invasive BCIs [2]. Non-invasive BCIs are the ones where the electrodes are simply attached to the scalp. In the case of Invasive BCIs, the electrodes are directly attached to the human brain. While Non-invasive BCIs are safe, invasive BCIs show better accuracy. The working of a BCI involves detecting and translating real-time brain signals that are a representation of the user’s intentions and displaying them on an output device as a final result. The four main steps [3] to accomplish this task are the signal acquisition process, feature extraction process, feature interpretation/translation process, and displaying the output, which is represented in figure 1.

Acquisition– The most common signals are electric signals. These signals are acquired from the brain and then amplified and tuned, and noise is removed to convert the input signal into a format that can be digitized and sent to a system that can be further interpreted.

Feature extraction– This step involves extracting certain characteristics and footprints that indicate the presence or absence of a particular intent. These characteristics can be time-triggered EEG or ECoG response amplitudes and latencies, power within specific EEG or ECoG frequency bands, or firing rates of individual cortical neurons, etc.

Feature interpretation/translation – In this step, the characteristics obtained in the previous steps are analyzed, and a conclusion is made as to what may be the intent of the user. Since the human brain is extremely fast and complex, these translation algorithms must be dynamic and adaptive to new signals and features in real-time.

Output – Finally, the interpreted message is executed on an output device that displays the result of the aforementioned process. The output can be experienced in various forms like visual outputs, including moving of cursor on the screen or changing of channel, audio outputs may change the volume of a device, etc.

Challenges faced to implement BCI to real-world technologies [4-6]:

1. Information transfer rate – Since the interpretation of messages is complicated and in real-time, it is important for the input messages to be collected and transferred at an extremely high rate. However, even the highest average information transfer rate doesn’t match up with minimal requirement.

2. High error rate – Since the brain is a dynamic, complex organ, it is difficult to interpret many signals altogether. This complexity worsens in disabled users whose brains send highly variable signals due to fatigue, medication, or pre-existing medical conditions such as seizures and spams. All these factors lead to a high error rate. Since BCI is proposed to be a solution for people whose only way of communication is the interface, even the slightest error can be fatal in some situations. Hence, it is very important to devise algorithms that take into consideration these factors. These algorithms much also are developed in such a way that they prevent errors.

3. Autonomy – Since the Brain Computer Interface is being aimed primarily at severely dependent people, it should also provide them complete control of its operation. The switching on and off of the setup without external help is a problem. Furthermore, the setup of the electrodes in both invasive and non-invasive BCIs is a tassel. Thus, it is highly unlikely that a disabled person can handle the system autonomously, making them dependent once again.

4. Environment and setup – All the developments that have been made in this field are made under the assumption of the environment being a laboratory or confined environments. The BCI setup is quite elaborate and is difficult to more from one location to another. Also, as the brain moves from a quitter place to a busy one, the signals become more complicated and unreadable. Hence, the environment should as be a factor while developing the controls for a BCI system.

Future goals:

The priority of researchers and scientists at the moment is to generate BCI solutions that use safer methods that provide better accuracy. This can be achieved by using protected, better-quality sensors that are safe for the user. By using the latest developments, the feature extraction and translation techniques will get more refined and defined, and hence the accuracy of the output will increase.

References:

[1] Wolpaw, Jonathan R., et al. “Brain-computer interface technology: a review of the first international meeting.” IEEE transactions on rehabilitation engineering 8.2 (2000): 164-173.

[2] Belkacem, Abdelkader Nasreddine, et al. “Brain computer interfaces for improving the quality of life of older adults and elderly patients.” Frontiers in Neuroscience 14 (2020): 692.

[3]Shih JJ, Krusienski DJ, Wolpaw JR. Brain-computer interfaces in medicine. Mayo Clin Proc. 2012;87(3):268-279. doi:10.1016/j.mayocp.2011.12.008

[4] Martini, M. L., Oermann, E. K., Opie, N. L., Panov, F., Oxley, T., & Yaeger, K. (2020). Sensor modalities for brain-computer interface technology: a comprehensive literature review. Neurosurgery, 86(2), E108-E117.

[5] Bockbrader, M. (2019). Upper limb sensorimotor restoration through brain–computer interface technology in tetraparesis. Current Opinion in Biomedical Engineering, 11, 85-101.

[6] Lotte, F., & Roy, R. N. (2019). Brain–computer interface contributions to neuroergonomics. In Neuroergonomics (pp. 43-48). Academic Press.

Cite this article:

A. Sethi, K. Yadav, Konstantinos Psannis (2021). Brain Computer Interface Technology. Insights2Techinfo. pp. 1

99930cookie-checkBrain Computer Interface Technologyyes