Neurotechnology and Brain Computer

Mohammed Sufyaan Hemang Gehlot Mohammed Aatif

Dept. of Electronics & Dept. of Electronics & Dept. of Electronics &
Communication Engineering Communication Engineering Communication Engineering
T John Institute of T John Institute of T John Institute of
Technology Technology Technology
[email protected]

Astract: Neurotechnology and brain- The research landscape in neurotechnology and

computer interfaces (BCIs) have emerged as BCIs is rapidly evolving. Recent advancements
cutting-edge interdisciplinary fields that include:
converge neuroscience, engineering, and
computer science. These technologies hold 1. Improved signal processing techniques
significant potential for revolutionizing human- for EEG and fMRI data, enabling more
machine interaction and therapeutic accurate and reliable brain-computer
communication.
interventions.
Neurotechnology encompasses advanced tools 2. Development of wearable devices that
and techniques for interfacing with the nervous can record and transmit brain activity
system. This field aims to develop non- data in real-time, paving the way for
invasive methods for monitoring and more convenient and accessible BCIs.
modulating brain activity. Examples of 3. Advances in machine learning
neurotechnology applications include brain- algorithms that can accurately classify
computer interfaces (BCIs), functional and interpret brain activity data,
magnetic resonance imaging (fMRI), and enabling the development of more
electroencephalography (EEG). intelligent and adaptive BCIs.
Brain-computer interfaces (BCIs) represent a 4. Increasing awareness of the ethical,
subset of neurotechnologies that aim to legal, and societal implications of
establish direct communication pathways neurotechnology and BCIs, driving the
between the brain and external devices. BCIs development of guidelines and
have the potential to revolutionize the way regulations to ensure responsible and
humans interact with computers and other safe use of these technologies.
technologies. Applications of BCIs include The potential societal impact of advancements
assistive devices for individuals with physical in neurotechnology and BCIs is immense.
disabilities, cognitive enhancement tools for These technologies have the potential to
healthy individuals, and advanced medical enhancehumancognition, impr
diagnostics and treatments. o v e accessibility for individuals with
disabilities, and revolutionize medical
diagnostics and
treatments. However, it is crucial to carefully speed and accuracy; however, they may not be
consider the ethical, legal, and societal suitable for patients with significant head or
implications of these advancements to ocular motor impairments due to their reliance
ensure their responsible and safe use. on activities like gaze shifting.
I. INTRODUCTION In a study utilizing SSVEPs, two superimposed
images with oscillating vertical and horizontal
Brain-Computer Interface (BCI) represents lines at different frequencies were employed as
cutting- edge technology enabling the visual stimuli. Offline analysis revealed that 7
translation of brain signals into predetermined out of 14 subjects successfully utilized the
commands for communication with individuals system for binary selection, with predicted
or the control of external devices. BCI accuracies ranging around 60-70% for most
researchers leverage various noninvasive participants.
neural signals, including Another SSVEP-based BCI approach involves
electroencephalography (EEG the use of rotating sets of dots with different
) , magnetoencephalography ( MEG), colors and flicker frequencies, inducing the
and functional magnetic resonance perception of two superimposed, transparent
imaging (fMRI). surfaces. Users control the BCI system by
EEG stands out as the most commonly utilized selectively attending to one of the two surfaces.
method due to its noninvasiveness, high
temporal resolution, portability, and reasonable II. SURVEY
cost. In our survey, we examined the number
of published BCI articles across different EEG-Based Brain-Computer Interfaces (BCIs):
neural signal recording modalities. Paradigms and Feature Types
Additionally, we delved into various aspects of
EEG-based BCI articles, including BCI A variety of paradigms have been utilized to
paradigms, objectives, applications, feature develop EEG-based BCI systems. EEG-based
types, classification algorithms, system types BCI articles were categorized into seven
(offline or online), and the nationalities of groups based on the experimental paradigms
corresponding researchers. employed: motor imagery, visual P300, steady-
BCIs establish a direct communication state visual evoked potential (SSVEP),
pathway between the human brain and the nonmotor mental imagery, auditory, hybrid,
external world, bypassing normal motor output and other paradigms. Hybrid paradigms, which
pathways. They offer an alternative means for simultaneously employ multiple BCI
individuals with severe motor disabilities yet p a r a d i g m s , h a v e e m e rg e d r e c e n t
intact cognitive abilities to interact with their l y, demonstrating increased interest in
environment. combining approaches for enhanced BCI
performance.
One of the major BCI paradigms employs
modulation of steady-state visual evoked For instance, some studies have integrated
potentials (SSVEPs). In a typical SSVEP BCI visual P300 and motor imagery paradigms to
system, multiple stimuli flickering at different control brain- controlled wheelchairs,
frequencies are presented to the user, who showcasing the potential of hybrid approaches
overtly directs attention to one stimulus by (Rebsamen et al., 2010). Other paradigms like
changing gaze direction. This results in "non-motor mental imagery" and "auditory"
enhancedSSVEP responses at t have been intermittently used in EEG-based
h e corresponding frequency over occipital BCI research. Additionally, novel paradigms
brain areas. SSVEP BCIs demonstrate based on covert attention, motion-onset visual
commendable evoked potentials (mVEPs), flash onset and
offset visual evoked potentials (VEPs), and
error potentials have been proposed. Subjects:
Recent advancements include the introduction The experiment involved 22 subjects,
of SSVEP-based mental spelling systems and comprising 11 graduate students from
the development of BCI applications for Tsinghua University, China (2 females and 9
navigation using wheelchairs, virtual reality, males), and 11 graduate students from
and robot arm controls. Notably, BCI University Medical Center Hamburg-
applications for diverse purposes, such as Eppendorf (UKE), Germany (5 females and 6
brain-controlled smart home systems, mobile males). Participants were aged between 20 and
phone applications, and real-time drowsiness 35 years and had normal or corrected-to-
detection systems, have seen significant normal vision. Four subjects, two from each
growth, highlighting the expanding utility of location, were excluded from data analysis due
BCI technology. to color blindness or failure to evoke SSVEP
at one of the stimulation frequencies. Thus, a
Feature Types in BCI Research total of 18 subjects were included in the study.

Various feature types have been employed in Stimuli:

BCI research to discriminate user intentions Stimuli were presented on an LCD monitor
accurately. Feature types were classified into with a refresh rate of 60 Hz and a viewing
five groups: power spectral density (PSD), distance of 60 cm. The stimuli consisted of two
event- related potential ( ERP), phase sets of dots (blue and red) randomly distributed
information, use of more than two feature in an annular area around a central fixation dot.
types, and others. Traditionally, PSD has been Each dot subtended 0.3° of visual angle. The
the most commonly used feature type due to its blue dots flickered continuously at 10 Hz, and
modulation by specific mental tasks across the red dots at 12 Hz, with accompanying
different frequency bands. rotational motion inducing the perception of
Among ERPs, P300 has been extensively two transparent, superimposed surfaces.
utilized, with a gradual increase in the adoption
of other ERPs such as N100, N200, P100, Experimental Procedure:
P200, movement-related ERP, and error- Each trial began with a cue indicating which
related ERP. Some studies have combined surface to attend, followed by the presentation
multiple feature types to enhance BCI of the stimulus for 4 s. Subjects were instructed
performance, while others have explored to direct their attention to the designated
various feature types independently to surface while fixating on the central white dot.
validate signal processing algorithms. Feedback about the recognized brain state was
In summary, EEG-based BCIs encompass a provided after each trial. The experiment
diverse range of paradigms and feature types, utilized a 3-day training paradigm, with each
reflecting ongoing advancements and day comprising training sessions followed by
innovations in the field. The integration of online testing sessions.
hybrid paradigms and the exploration of novel
feature types hold promise for further EEG and EOG Recording:
improving BCI performance and expanding its EEG signals were recorded using a 32-channel
applications. EEG amplifier with a sampling rate of 128 Hz.
Electrodes were positioned according to the
10–20 system, with additional electrodes used
III. METHODS to record horizontal and vert
ical
electrooculography (EOG) for eye movement unsatisfactory, the training dataset and the
detection in nine subjects. previous classifier were retained. This
approach allowed for gradual augmentation of
Analysis Methods the sample size to enhance performance and
SSVEP Feature Extraction: Canonical stability while excluding sessions with poor
Correlation Analysis (CCA): subject performance.
To address inter-subject variability, canonical
correlation analysis (CCA) was employed for Classifier Parameter Updates
automatic channel selection. CCA measures While the classifier parameters were adjusted
the linear relationship between from session to session within each training
multidimensional variables and identifies basis day, they were not carried over from one day to
vectors that maximize the correlation between the next. Subjects commenced each training
variable projections. A parameter-free CCA- day with a new session to assess adaptation to
based algorithm was utilized, showing the BCI system. This design choice aimed to
improved performance compared to existing explore human adaptation to the BCI system
SSVEP BCI systems. independently of parameter transfer between
In summary, the experimental methods days, thereby focusing on the accumulation of
involved subject recruitment, stimulus human experience with the BCI system over
presentation, experimental procedures, EEG time.
and EOG recording, and analysis methods
focusing on SSVEP feature extraction using IV. INFERENCE
CCA. These methods were implemented to
assess cross-laboratory reproducibility and The culmination of our extensive literature
reliability of results in an SSVEP-based BCI surveyand research endeavors
paradigm. i n Neurotechnology and Brain-
Enhanced Correlation and Feature Vector Computer Interfaces (BCIs) yields several
Formation noteworthy inferences:
A stronger correlation between the EEG signal
(E) and the reference signal (R) is anticipated 1. Advancements in Neural Signal Recording
at the corresponding frequency, resulting in Modalities:
increased correlation coefficients (ρ). These Our investigation reveals a notable shift
coefficients were computed for each trial. To towards noninvasive techniques, particularly
prevent overtraining, only the two largest EEG, due to its high temporal resolution,
coefficients at each stimulation frequency were portability, and cost-effectiveness. This
utilized to construct a four-dimensional feature indicates a growing preference for accessible
vector for classification. Figure 2(a) illustrates technologies in BCI development.
the utilization of the CCA method within our 2. Emerging Trends in EEG-based BCI
algorithm. Research:
Recent BCI studies predominantly integrate
Classification Strategy EEG technology, reflecting a surge in global
For each training session, if the classifier was publications across diverse scientific domains.
deemed effective (based on performance with This trend underscores EEG's recognized
the current dataset), the data from the current potential for advancing neurotechnology and
session were incorporated into the training fostering interdisciplinary collaborations.
dataset, and the updated classifier was 3. Technological Innovations:
employed for subsequent sessions. Conversely, Reviewed literature emphasizes simultaneous
i f t he c l a s s i f i e r ' s p e r f o r m a n c e focus on hardware advancements, including
was
microelectrode arrays, neural signal processing understanding brain function. A comprehensive
circuits, and wearable brain caps. This analysis of recent trends and developments
underscores the ongoing dedication to reveals a multifaceted landscape characterized
enhancing BCI system efficiency and by increasing publication rates, paradigm
practicality through hardware innovation. diversification, and a growing emphasis on
4. Ethical Considerations and Societal practical applications for individuals with
Implications: neuromuscular disorders.
A significant subset of literature explores One of the most striking trends in EEG-based
ethical concerns surrounding BCI research, BCI research is the steady rise in the number of
extending beyond technological progress to published articles from 2007 to 2011. This
broader societal issues like privacy, informed exponential growth reflects the escalating
consent, and the ethical impacts of BCIs on interest and investment in BCI technology
individuals and communities. across academia, industry, and healthcare
5. Challenges and Opportunities: sectors. Researchers and practitioners alike are
Key challenges, such as motor impairments, drawn to the promise of BCIs as a means of
highlight the necessity for adaptive BCI facilitating direct communication and control
systems. The introduction of novel paradigms, for individuals with severe motor disabilities,
like SSVEP modulation, presents opportunities offering a lifeline to those who may have
to address these challenges and broaden BCI otherwise been marginalized by their physical
applications. limitations.
6. Gaps in Literature:
While literature reviews offer valuable While motor imagery-based BCI paradigms
insights, a noticeable gap exists in have traditionally dominated the field, recent
comprehensively exploring general trends years have seen a notable diversification in
specific to EEG-based BCIs. Our study aims to experimental approaches. In particular, there
fill this void by examining publication trends, has been a surge in the utilization of steady-
characteristics, and novel applications within state visual evoked potential (SSVEP) and
the EEG-based BCI landscape. visual P300 paradigms. These paradigms
In conclusion, the inferences derived from our leverage the brain's response to visual stimuli,
literature survey and research provide a offering distinct advantages in terms of ease of
foundation for deeper comprehension of the use, robustness, and signal-to-noise ratio. The
evolving landscape of Neurotechnology and growing popularity of SSVEP and visual P300
BCIs. These insights enrich ongoing discourse BCIs underscores the importance of
in the field and serve as a springboard for developing versatile and user-friendly
further research and technological progress. interfaces that cater to a diverse range of user
needs and preferences.
V. CONCLUSION A key driving force behind the proliferation of
EEG-based BCI research is the increasing
focus on practical applications for individuals
Over the past decade, EEG-based Brain- with neuromuscular disorders. As the field
Computer Interface (BCI) research has matures, researchers are shifting their attention
witnessed remarkable growth and innovation, towards developing BCI systems that can be
fueled by advancements in neuroscience, seamlessly integrated into everyday life,
technology, and computational methods. This offering tangible benefits to end-users. This
burgeoning field holds tremendous potential paradigm shift is reflected in the growing
for revolutionizing communication, assistive number of studies aimed at developing
t echnology, n e u r o r e h a b i l i t a t i o n , and practical BCI paradigms and applications,
ranging from communication aids and underscore the transformative potential of BCI
neuroprosthetics to assistive devices and technology and pave the way for future
neurorehabilitation tools. innovations in assistive technology,
The methodological approach employed in neurorehabilitation, and beyond. As
recent studies underscores a commitment to researchers continue to push the boundaries of
rigor and innovation in EEG-based BCI what is possible, EEG-based BCIs are poised to
research. Thorough literature surveys and trend become indispensable tools for unlocking the
analyses provide valuable insights into the full potential of the human mind.
evolving landscape of BCI technology,
shedding light on emerging paradigms, VI. REFERENCES
classification algorithms, and experimental
methodologies. By systematically evaluating [ 1 ]
the prevalence of different experimental WolpawJR,BirbaumerN,McFarlandDJ,Pfurtsch
paradigms, research aims, and classification ellerG and Vaughan T M 2002 Brain–computer
algorithms, researchers can identify trends, interfaces for communication and control Clin.
gaps, and opportunities for future exploration. Neurophysiol.
Furthermore, recent studies have demonstrated
the efficacy and feasibility of EEG-based BCI
systems in real-world settings, underscoring [2] Zhang, D., Maye, A., Gao, X. R., Hong, B.,
Engel, A. K., & Gao, S. K. (2010). An
their potential as transformative tools for
independent brain–computer interface using
individuals with neuromuscular disorders. For
covert non-spatial visual selective attention.
example, a novel SSVEP-based BCI system, Journal of Neural Engineering, 7, 16010.
developed using covert non-spatial visual
selective attention, showed promising results in [3] Wolpaw, J. R., Birbaumer, N., McFarland,
online training programs with healthy subjects. D. J., Pfurtscheller, G., & Vaughan, T. M.
The study revealed significant improvements (2002). Brain–computer interfaces for
in control accuracy over the training period, communication and control. Clinical
highlighting the potential of SSVEP-based Neurophysiology, 113, 767–791.
BCIs as a means of enhancing communication
[4] [29]
and control for paralyzed patients.
McFarlandDJ,McCaneLM,DavidSVandWolpa
wJR 1997 Spatial filter selection for EEG-ba
In conclusion, the burgeoning field of EEG-
based Brain-Computer Interface (BCI) research
holds immense promise for advancing our
understanding of the brain, enhancing human-
computer interaction, and improving the
quality of l i fe for individuals with
neuromuscular disorders. The recent trends and
developments outlined in this analysis