The role of artificial intelligence in EEG Biometric recognition
RASHA AHMED JELAB1*,Zadoon Tark abdlwahab1*
1Tikrit University, College of Computer Science and Mathematics, Computer Science Department, Tikrit, Iraq,
Email: rasha.a.jelab22ms@st.tu.edu.iq
Email: zaidoon@gmail.com
ABSTRACT:
Artificial Intelligence technology is invading human life with its numerous leverages. Like all other fields, EEG biometric recognition also adopted AI tools and techniques to enhance subject identification. The role of AI in EEG biometric recognition is widespread, covering identification, authentication, and administrative features. In this research, we carefully discuss the role of AI in EEG biometric recognition. We discussed the superior AI technologies and briefly described each that can play a vital role in biometric recognition. Combining brain science research with artificial intelligence technology makes the brain’s electrical Characteristics expected to become new biometrics in identification. Exploiting brain signals for biometric recognition has received significant attention from the scientific community in the last decade, with most of the efforts so far devoted to the quest for discriminative information within electroencephalography (EEG) recordings.Biometric recognition systems are nowadays deployed in many real-life applications,ranging from border control to verification on a mobile device. Although several biometric traits have already achieved the plateau of their productivity, such as EEG signals, fingerprints, and iris, prominent research is still carried out to design innovative recognition systems, relying on other characteristics that may possess useful properties not available in mainstream solutions. In this regard, cognitive biometrics are used in addition to using physical or behavioural traits. The need for a biometric is incontrovertible. Additional requirements for a biometric system are security, capability of meeting reliability, accuracy and adaption standards. Authentication would require a living cooperative person, as opposed to inanimate records that can be forged or passive biometric identifiers that can be extracted from someone who is forced or even dead. The biometric system measures individuals’ unique physical or behavioural characteristics to recognize or authenticate their identity.
Keywords:)Artificial intelligence, Brain-Computer Interface (BCI), Biometric, Electroencephalography (EEG), Motor Imagery(.
1. Introduction
Artificial intelligence (AI) is a broad area of computer science that makes machines function like a human brain. It is used to address issues that are difficult to clarify using traditional computational techniques[1]. Artificial Intelligence technology is invading human life with its numerous leverages. Like all other fields, education also adopted AI tools and techniques to enhance student learning capabilities and other features of the education sector. Artificial Intelligence (AI) gains rapid response in all aspects of human life. Day-to-day life has changed after the wide-spreading adaptability of AI. Using AI tools and techniques, researchers and AI scientists are making their unstinting efforts to make human life more comfortable and technologically advanced[2]. The aim of AI is to facilitate systems with intelligence that is capable of human-like learning and reasoning. It possesses tremendous advantages and has been successfully applied in numerousindustrial areas, including image classification, speech recognition, autonomous cars, computer vision, etc. With immense potential, power electronics benefit from the development of AI[3]. AI research’s central problems (or goals) include reasoning, knowledge, planning, learning, natural language processing (communication), perception and the ability to move and manipulate objects. General intelligence is among the field’s long-term goals. Approaches include statistical methods, computational intelligence, and traditional symbolic AI. Many tools are used in AI, including versions of search and mathematical optimization, logic, and methods based on probability and economics. The AI field draws upon computer science, mathematics, psychology, linguistics, philosophy, neuroscience, artificial psychology, plus many others[4]. Biometrics such as EEG, fingerprint and iris scans are being utilized to authenticate workplace employees and identify smartphone owners. Such biometrics can be implemented in organizations to authorize data access for confidential data. Biometrics can be used along with traditional passwords or PIN numbers for multi-factor authentication. Additionally, the adoption of AI will help develop data-driven security protocols. Hence, clubbing AI and biometrics together will create a dynamic security model. systems can minimize’ human error ‘ events, provided they are properly programmed and can contribute to making quicker choices through cognitive techniques. It is considered a very effective technology as it challenges hackers to penetrate and ensures user safety[5]. Applications include determining how to make machines use language, form abstractions and concepts, and solve complicated problems. The massive and growing data available today and the steady improvements in computational power and algorithms have generated numerous applications of AI across many diverse industries. Although definitions and concepts of AI vary with the goals and domain, the main characteristic of AI is mimicking human cognitive function, particularly learning and problem-solving. Notably, the concept proposed in 1956 is still relevant to current AI research[6].
2. Brain-Computer Interface(BCI)
The contemporary era demands progress with respect to manual work or even semi-machine dependence, and the desired procession can be provided by Brain-Computer Interface (BCI). As the name suggests, BCI is a bridge between the signals that are generated by the thoughts in our brain and the machine that can implement the produced signals into actions. It is a breakthrough invention in the field of brain-mapping science which can adequately help and aid impaired vision, movement, hearing and any damaged functioning of the body that can be thought of as practically[7]. Brain-computer interface (BCI) technology is a powerful communication tool between users and systems. It does not require any external devices or muscle intervention to issue commands and complete the interaction [8]. It helps the subject become more independent as it provides them with complete confidentiality. Modern BCIs use electroencephalographic (EEG) activity recorded at the scalp or single-unit activity recorded from within the cortex to control cursor movement, pick letters or icons, or operate a neuro-prosthesis[9]. Electroencephalogram (EEG) signals extracted during imagined activities have been studied for use in Brain-Computer Interface (BCI) applications. The major hurdle in the EEG-based BCI is that the EEG signals are unique to each individual. This complicates a universal BCI design. On the contrary, this disadvantage is the advantage when it comes to using EEG signals from imagined activities for biometric applications[10]. In recent years, it has been shown that EEG is an appropriate signal for biometricauthentication and has important features such as resistance to spoofingattacks and impossibility to use under pressure and coercion states[11]. Brain computer interface (BCI) systems are used in a wide range of applications such as communication, neuro-prosthetic and environmental control for disabled persons using robots and manipulators[12].
3. Biometrics
Biometrics as a terminology refers to the combination of the Greece words (Bio) and (Metrics), which means ‘’life measurements’’[13]. Also, Biometrics could be defined as the science and technology of measuring and analyzing biological data. In information technology, biometrics refers to technologies that measure and analyses human body characteristics, such as DNA, fingerprints, eye retinas and irises, voice patterns, facial patterns and hand measurements, for authentication purposes.
Biometrics could be divided into 2 major classes:
1.1 Physical biometrics These involve some form of physical measurement and include modalities such as face, fingerprints, iris scans, hand geometry, etc.
1.2. Behavioural biometrics These are usually temporal in nature and involve measuring the way in which a user performs certain tasks. This includes modalities such as speech, signature, gait, keystroke dynamics, etc[14].
Biometric Systems
Biometric systems are, in fact, systems of pattern recognition which operate via the acquisition of biometrical data from a person. Obtaining a set of features from data that has been obtained and making a comparison of this set of features with the set of templates in the database. According to the context of the application, biometric systems would operate in either mode of identification or verification.
· In the Mode of Identification, the system can recognize a person by searching all users’ templates in the database for a match. This is why this system carried out a one-to-many comparison that aims at establishing the identity of a person.
· In the Mode of Verification, the system can validate the identity of an individual by conducting a comparison of the obtained biometrical data with their own biometrical template (or templates), which are stored earlier in the database of the system[15].
Figure (1); user enrolment that is a common part of each one of the abovementioned tasks, has been graphically represented as well[15] .
Figure (1): Block Diagrams of Enrollment, Verification and Identification Tasks[15]
Characteristics of Biometric Systems
Every biometric characteristic requires satisfying the following properties:
Universality: The characteristic should be encountered in each person. Uniqueness: The characteristic should be unique between individuals. Permanence: The characteristic should be sufficiently constant over long time. Collect ability: The characteristic can be measured from a quantity point of view. Performance: The characteristic gives an accurate result under different environment.
Acceptability: The entire people accept to give their traits to the system without any Problem.
Circumvention: The characteristic must be hard to deceive and imitate[14].
Applications of Biometric Systems
Biometric technology is used for variety of applications and can be divided into main three groups- Commercial, Government and Forensic applications. The biometrics is deployed in many areas shown in the following:
§ Commercial Applications: Access Control (Logical Access and Physical Access), Time and Attendance Management, Financial Services and Banks.
§ Government Applications: Border Control / Airports, Security and Immigration checks, Communication Systems, Healthcare and Social Services.
§ Forensic Applications: Justice / Justice / Law Enforcement, Surveillance[16].
4. Electroencephalography (EEG)
Electroencephalography (EEG) signals measure the post-synaptic brain activities by electrical potentials from the scalp using lightweight and non-invasive devices. Due to its high temporal resolution and rich dynamics, it is considered one of the most promising biological signals for biometric applications, among others, such as electromyography (EMG) and electrocardiography (ECG). Furthermore, EEG presents several peculiarities, making it more advantageous than traditional biometric modalities such as fingerprint and iris in terms of robustness against spoofing attacks, privacy compliance, and aliveness detection[17]. we study the feasibility of using Electroencephalography (EEG) brain signals for authentication purpose[18]. these signals are measured easily, specifically the EEG signal, using portable and inexpensive devices with dry electrodes. EEG signals were involved in non-medical devices such as brain-computer interfaces [19].
An electroencephalogram (EEG) is a widely used neuro-imaging modality which
can measure the potential changes in voltages caused by the firing of electrical
impulses of neurons, detected on the scalp and captured by a device containing a pattern of electrodes. EEG signals are categorized based on their frequency and divided into five major bands: delta band from 0.5 to 3 Hz, theta band from 4 to 7 Hz, alpha band from 8 to 13 Hz, beta band from 14 to 30 Hz, and gamma band greater than 30 Hz. Each of these spectra is generally associated with some kind of activity like moving fingers, sleeping, active thinking, or problem-solving. EEG is used in brain-computer interfaces (BCI), which allow the human subject and computer to communicate with no contact. Applications of such research are gaining popularity to the field of affective computing, which aims to understand the states of the human mind [20].
5. motor imagery (MI)
The MI protocol is designed to collect EEG signals corresponding to a series of right-arm, left-arm, right-leg and left leg imaginary movements[21]. MI is a common mental task in which subjects are instructed to imagine themselves performing a specific motor action (such as a hand or foot movement) without an overt motor output. There are various acquisition techniques for capturing MI brain activities. Among these techniques, EEG is the most studied measure of potential for non-invasive BCI designs, mainly due to its excellent temporal resolution, non-invasiveness, usability, and low set-up costs[22] MI EEG signals are brain activity recorded when the subject imagines or intends to perform actions like hand or leg movement. The MI EEG signal is produced in the brain’s sensorimotor cortex area as a response to these imagining or thinking tasks These MI signals have been utilized by researchers to discriminate between different oscillatory brain activations for different tasks. Automated MI classification has been performed using different machine learning and deep learning techniques.[23].
The robustness and computational load are the key challenges in motor imagery (MI) based on electroencephalography (EEG) signals to decode for the development of practical brain-computer interface (BCI) systems[24]. BCIs based on motor imagery have been intensively used in rehabilitation applications that seek to assist disabled individuals as well as restore individual’s physical and cognitive functions lost due to neural disorders. Motor imagery (MI) is the process of movement imagination without any actual muscle activation. It was found that the MI process activates the same brain regions activated during the actual physical movement. Therefore, during the rehabilitation process, patients with motor impairments practice the MI process to activate the injured brain motor regions[25].
The term motor imagery refers to imagination of moving a body part without actually moving it. This imagination causes ERD activities in brain that can be observed in EEG. However, ERD/ERS patterns of all body parts cannot be discriminated in EEG. The produced patterns should be large enough to be distinguished from the background EEG. Currently, there are four types of motor imagery actions that can be detected via EEG. These actions are movement of left hand, right hand, feet and tongue. These four motor imagery signals can be used to control BCI after attending sufficient training sections[26] .
The physiological mechanisms underlying the process of motor imagery have significant similarities with the mechanisms of motor control, and this can be used for the rehabilitation of patients with movement disorders. In patients with severe paresis, motor imagery may be the only method producing movement recovery[27].
There are various acquisition techniques for capturing MI brain activities. Among these techniques, EEG is the most studied measure of potential for noninvasive BCI designs, mainly due to its excellent temporal resolution, noninvasiveness, usability, and low set-up costs[22].
6. Conclusion
Artificial Intelligence has an impact on almost every field of life. EEG biometric recognition is one of the leading areas leveraged by AI technology. Like all other fields, biometric recognition also adopted AI tools and techniques to enhance subject identification. The role of AI in EEG biometric recognition is wide-spreading, covering identification, authentication, and administrative features. In this research, we carefully discuss the role of AI in EEG biometric recognition. In this research, we discussed the superior AI technologies and briefly described each that can play a vital role in biometric recognition. The integration of brain science research and artificial intelligence technology is anticipated to result in the utilization of the brain’s electrical characteristics as novel biometrics for identifying purposes. The scientific community has shown considerable interest in utilizing brain signals for biometric recognition during the past decade. The majority of research efforts have focused on identifying discriminative information within electroencephalography (EEG) recordings. Biometric recognition systems are currently utilized in several practical applications, encompassing border control and mobile device verification. While certain biometric traits, such as EEG signals, fingerprints, and iris, have reached a point of maximum effectiveness, ongoing research is being conducted to develop novel recognition systems that utilize alternative characteristics. These characteristics may offer unique properties that are not present in commonly used solutions. Cognitive biometrics are employed in conjunction with physical or behavioral characteristics in this context. The necessity of implementing a biometric system is indisputable. In addition to the aforementioned criteria, it is imperative for a biometric system to possess some supplementary requisites, namely, a robust security framework, the ability to fulfill stringent dependability, accuracy, and adaptability standards. The process of authentication necessitates the presence of an active and participating individual, as opposed to relying on inanimate records that are susceptible to forgery or passive biometric identifiers that can be obtained from an individual through coercion or even posthumously. The biometric system is designed to assess and verify the identity of individuals by analyzing their distinct physical or behavioral attributes.
[1] R. Abduljabbar, H. Dia, S. Liyanage, and S. A. Bagloee, “Applications of artificial intelligence in transport: An overview,” Sustainability, vol. 11, no. 1, p. 189, 2019.
[2] B. Namatherdhala, N. Mazher, and G. K. Sriram, “A comprehensive overview of artificial intelligence tends in education,” International Research Journal of Modernization in Engineering Technology and Science, vol. 4, no. 7, 2022.
[3] S. Zhao, F. Blaabjerg, and H. Wang, “An overview of artificial intelligence applications for power electronics,” IEEE Transactions on Power Electronics, vol. 36, no. 4, pp. 4633-4658, 2020.
[4] P. Ongsulee, “Artificial intelligence, machine learning and deep learning,” in 2017 15th international conference on ICT and knowledge engineering (ICT&KE), 2017: IEEE, pp. 1-6.
[5] A. P. Iyer, J. Karthikeyan, R. Khan, and P. Binu, “An analysis of artificial intelligence in biometrics—the next level of secu‑rity,” J Crit Rev, vol. 7, no. 1, pp. 571-576, 2020.
[6] J. Lee, T. Suh, D. Roy, and M. Baucus, “Emerging technology and business model innovation: the case of artificial intelligence,” Journal of Open Innovation: Technology, Market, and Complexity, vol. 5, no. 3, p. 44, 2019.
[7] R. A. Ramadan and A. V. Vasilakos, “Brain computer interface: control signals review,” Neurocomputing, vol. 223, pp. 26-44, 2017.
[8] J. R. Munavalli, P. R. Sankpal, A. Sumathi, and J. M. Oli, “Introduction to Brain–Computer Interface: Applications and Challenges,” Brain‐Computer Interface: Using Deep Learning Applications, pp. 1-24, 2023.
[9] B. K. Zaidi, A. Arshad, and A. Gupta, “Brain Computer Interface (BCI) Technology: A Neurological Restoration,” in Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2022), 2023: Springer, pp. 709-716.
[10] S. Bak and J. Jeong, “User Biometric Identification Methodology via EEG-Based Motor Imagery Signals,” IEEE Access, 2023.
[11] A. J. Bidgoly, H. J. Bidgoly, and Z. Arezoumand, “A survey on methods and challenges in EEG based authentication,” Computers & Security, vol. 93, p. 101788, 2020.
[12] M. Z. Baig, N. Aslam, and H. P. Shum, “Filtering techniques for channel selection in motor imagery EEG applications: a survey,” Artificial intelligence review, vol. 53, pp. 1207-1232, 2020.
[13] I. M. Alsaadi, “Physiological biometric authentication systems, advantages, disadvantages and future development: A review,” International Journal of Scientific & Technology Research, vol. 4, no. 12, pp. 285-289, 2015.
[14] J. Choudhary, “Survey of different biometrics techniques,” International Journal of Modern Engineering Research (IJMER), vol. 2, no. 5, pp. 3150-3155, 2012.
[15] A. K. Jain, A. Ross, and S. Prabhakar, “An introduction to biometric recognition,” IEEE Transactions on circuits and systems for video technology, vol. 14, no. 1, pp. 4-20, 2004.
[16] T. Sabhanayagam, V. P. Venkatesan, and K. Senthamaraikannan, “A comprehensive survey on various biometric systems,” International Journal of Applied Engineering Research, vol. 13, no. 5, pp. 2276-2297, 2018.
[17] M. Wang, J. Hu, and H. A. Abbass, “BrainPrint: EEG biometric identification based on analyzing brain connectivity graphs,” Pattern Recognition, vol. 105, p. 107381, 2020.
[18] T. Pham, W. Ma, D. Tran, P. Nguyen, and D. Phung, “A study on the feasibility of using EEG signals for authentication purpose,” in Neural Information Processing: 20th International Conference, ICONIP 2013, Daegu, Korea, November 3-7, 2013. Proceedings, Part II 20, 2013: Springer, pp. 562-569.
[19] M. Abo‐Zahhad, S. M. Ahmed, and S. N. Abbas, “State‐of‐the‐art methods and future perspectives for personal recognition based on electroencephalogram signals,” IET Biometrics, vol. 4, no. 3, pp. 179-190, 2015.
[20] V. Doma and M. Pirouz, “A comparative analysis of machine learning methods for emotion recognition using EEG and peripheral physiological signals,” Journal of Big Data, vol. 7, no. 1, pp. 1-21, 2020.
[21] R. Das, E. Maiorana, and P. Campisi, “Motor imagery for EEG biometrics using convolutional neural network,” in 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018: IEEE, pp. 2062-2066.
[22] S. Siuly and Y. Li, “Improving the separability of motor imagery EEG signals using a cross correlation-based least square support vector machine for brain–computer interface,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 20, no. 4, pp. 526-538, 2012.
[23] S. U. Amin, M. Alsulaiman, G. Muhammad, M. A. Mekhtiche, and M. S. Hossain, “Deep Learning for EEG motor imagery classification based on multi-layer CNNs feature fusion,” Future Generation computer systems, vol. 101, pp. 542-554, 2019.
[24] M. T. Sadiq et al., “Motor imagery EEG signals decoding by multivariate empirical wavelet transform-based framework for robust brain–computer interfaces,” IEEE access, vol. 7, pp. 171431-171451, 2019.
[25] A. Khalaf, E. Sejdic, and M. Akcakaya, “A novel motor imagery hybrid brain computer interface using EEG and functional transcranial Doppler ultrasound,” Journal of neuroscience methods, vol. 313, pp. 44-53, 2019.
[26] A. Schlögl, F. Lee, H. Bischof, and G. Pfurtscheller, “Characterization of four-class motor imagery EEG data for the BCI-competition 2005,” Journal of neural engineering, vol. 2, no. 4, p. L14, 2005.
[27] O. Mokienko, L. Chernikova, A. Frolov, and P. Bobrov, “Motor imagery and its practical application,” Neuroscience and Behavioral Physiology, vol. 44, pp. 483-489, 2014.




