Brain-Computer interfaces or BCIs are booming, where new equipment and analysis techniques are seen every day for applications development.
Imagine a personal laboratory where, through a simple headband and a smartphone, you can record your brain waves, extract significant patterns automatically and apply that information to the activities or tasks you perform everyday.
Through the combination of EEG sensors with smartphones, computers or tables, we offer new opportunities to obtain daily activity brain records. This is facilitated by the development of tools to interpret data and provide meaningful measures.
We intend to interpret brain activity quickly and efficiently to interact directly with hardware and software. It is common to work with motor activity to move a robotic arm, or to use brain information related to what a person wants to try to communicate without having the need to speak. Those are some of the applications of this type of technology.
Currently in Naxon we are developing software that will allow us to obtain significant patterns in an automated way.
Imagine being able to see how your brain responds to stimuli personally, with a laboratory on your smartphone. What until recently was extremely expensive, with the new technology and developments such us our work, we are making it possible for such applications to be available massively.
The development of portable systems and applications of brain computer interfaces is at the forefront of technological innovation. Naxon introduces a key area and presents new paradigms to the field of technology. This was demonstrated in the openness and interest we received during the summit. We hope to continue improving our technology and open up business opportunities and knowledge sharing.
The objective of the forum in which we participated is to provide an opportunity for participants to exchange the latest trends in the global services sector and identify business and investment opportunities for SMEs in Latin America and the Caribbean. More than 600 companies participated in the event.
As mentioned at mitagds, “the exponential acceleration of digital transformation presents a challenge and an opportunity for governments and companies in the Latin American and Caribbean region to unite to improve the lives of their citizens and the competitiveness of their economies … “
…I think what’s gonna happen tomorrow with Uber, you never now… In my view you will not even gonna be using a mobile phone by the way, to call a taxi. You will just think about it, validate it and it will come. Naxonlabs here in the city itself is writing or doing something on mapping the brain waves and analyzing it. So companies like that will make the Uber experience even different within the 2-3 years itself…
Global Digital Services Summit Buenos Aires – Avinash Vashistha (Chairman & CEO Tholons)
During the event we had the opportunity to establish contacts with various potential partners and investors, generating a first instance of dialogue for the development of common projects, opening the possibility of applying our technology to various areas of interest, at the same time as we get involved in other markets that generate new possibilities. Several projects were discussed and some are already underway.
One of the areas of gradual development of Naxon refers to applications that evaluate the quality of sleep and can be used personally and individually by users of portable EEG devices and professionals in the areas of Clinical Psychology and Sleep Medicine. The advantages of applying this technology to this area are multiple, for example, remote monitoring and better patient follow up. At Naxon, through the development of our integrated software platform with pattern analysis tools, we seek one of the applications to be the staging of sleep and the completion of preliminary analysis of the person’s sleep quality. Below we show the current problems in the area and how this new technology is introduced to it.
Sleep disorders in today’s world
Sleep disorders are one of the most relevant health problems currently. The importance of a good quality of sleep is not only fundamental as a determinant factor of health, but as an element that promotes a good quality of life (Sierra, Jiménez-Navarro, & Martín-Ortiz, 2002).
Sleep disorders refer to a wide variety of conditions, such as insomnia disorders, sleep-related breathing disorders, central disorders of hypersomnolence, circadian sleep-wake rhythm disorders, sleep-related movement disorders and parasomnias, among others (Sateia, 2014).
In a sample from the United States, a general prevalence of current or previous sleep disorders in adults of 52.1% was found. Specifically, the results show a prevalence of insomnia of 42.5%, 11.2% of nightmares, 7.1% of excessive sleep, 5.3% of somnolence and 2.5% of sleepwalking. These conditions were often chronic and usually started at an early age (Bixler, Kales, Soldatos, Kales, & Healey, 1979).
The sleep problems of older people are different from those of younger people. Sleep apnea and periodic leg movements (PLMS) during sleep are extremely common in older people. Complaints of insomnia, snoring, daytime sleepiness and lack of sleep, which are also very common in this population, should not be taken lightly, as they can be symptoms of these physiological sleep disorders. Also, subjective reports of lack of sleep, insomnia, snoring and excessive daytime sleepiness should not be taken lightly and should not be assumed to be a normal sign of aging (Ancoli-Israel, 1989).
It has been estimated that among people over 65 years of age, half of those who live in their own homes and 60% of those who live in residences suffer from a sleep disorder, of which , the difficulty to start and maintain sleep and Daytime sleepiness are more common. On the other hand, in the United States, 25 million prescriptions of hypnotics are extended every year and 40% of them are used by the elderly. In the case of Spain, 58% of people who consume hypnotics are elderly and, as is observed in other countries, the recommendations for their use are incorrectly followed, thus, drugs with intermediate or long-term action are used inappropriately. (Benetó, 2000).
On the other hand, approximately 25% of all children experience some type of sleep problem at some point during childhood. Several studies have examined the prevalence of sleep complaints reported by parents and children in large samples of healthy children and adolescents and with typical development, many of these have also delineated the association between interrupted sleep and behavioral concerns. Sleep problems are even more frequent in children and adolescents with chronic medical conditions, neurological and psychiatric development issues (Owens, 2007). Even in healthy children, a prevalence of 37.4% of sleep disorders was found. 40.7% of these children have problems such as resistance to sleep/fear of sleeping alone and between parasomnias, nightmares and night terrors are present (Convertini, Krupitzky, Tripodi, & Carusso 2003).
On the other hand, syndromes such as attention deficit hyperactivity disorder (ADHD) and sleep disorders have a high correlation and are very frequent in a neuropediatric clinic. By improving the sleep of these children, the symptoms of ADHD diminish and the administration of psychostimulants whose undesirable side effects generate great anxiety in the parents of these children is avoided (Fursow, Betancourt, & León 2006).
On the other hand, several studies have shown the high prevalence of sleep disorders among adolescents. As a cause, psychological, hormonal, puberal factors and inadequate sleep habits are invoked. In a sample of 1,155 students, 38.55% reported poor quality sleep, 23.1% difficulty in conciliation, 38.2% night awakenings and 15.9% early awakening. 17.7% reported a complaint of sleep plus some symptom of insomnia; the prevalence of insomnia found was 9.9%. Snoring (20.5%), Somnilochia (45.4%) and nightmares (29.5%) are the most frequent parasomnias. Additionally, 53% complain of excessive daytime sleepiness (Jiménez et al., 2004).
In university students, certain results show that approximately 30% have poor sleep quality, excessive latency and poor sleep efficiency; no differences were found between men and women in any component, with the exception of hypnotics use, where women had a higher score (Sierra, Jiménez-Navarro, & Martín-Ortiz, 2002).
In addition, hospitalized adult patients often have sleep problems. Most studies have focused mainly on patients in the intensive care unit. However, many medical and surgical disorders can alter the normal architecture of sleep (Aguilera Olivares, Díaz, & Sánchez, 2012).
The direct and indirect costs of sleep disorders are high. An analysis of this was carried out in Australia. The total financial cost (independent of the cost of suffering) of $ 4524 million represents 0.8% of the Australian gross domestic product. The cost of suffering of $ 2.97 billion is 1.4% of the total burden of the disease in Australia (Hillman, Murphy, Antic, & Pezzullo, 2006).
The study of sleep in the laboratory is carried out through the polysomnographic record (PSG), which consists of the simultaneous recording of the EEG, the electrooculogram (EOG), the electromyogram (EMG) of the submental muscles, the electrocardiogram (ECG) and breathing. In addition, other parameters can be monitored: rectal temperature, blood pressure, limb movements, blood gases, endosophageal pressure, penile erection and electrodermal reaction (Peraita-Adrados, 2005).
PSG measures the cycles and stages of sleep, either rapid eye movement (REM) sleep or synchronized sleep, or non rapid eye movement (NREM) sleep or desynchronized sleep, which has three stages that can be detected by means of the brain waves (EEG).
Polysomnography can be carried out both in a specialized sleep clinic and in the person’s own home. At a specialized center, you are asked to arrive approximately 2 hours before bedtime. The person will sleep in a bed in this center. The exam is often done during the night. Electrodes are placed on the chin, scalp and on the outer edge of the eyelids. There are monitors to record the heart rate and breathing attached to the chest, which remain in place while you sleep. The electrodes record signals while one is awake (with closed eyes) and during sleep. In the exam, the amount of time it takes to fall asleep is measured, as well as the time it takes to enter the rapid eye movements sleep. A trained specialist observes while one is asleep and notes any change in heart or respiratory rate. There are also monitors that record movements during the session. Sometimes a camcorder records movements during sleep. Due to the particular circumstances of doing it in a specialized center, with all the monitors and register equipment required, users usually report some difficulty to be able to fall asleep. In the case of doing it at home one can use a polysomnograph to help diagnose sleep apnea. In a sleep center the device is picked up or a specialized therapist comes to your home to prepare that device.
Home exams may be used if the sleep specialist thinks that you have obstructive sleep apnea and there are no other sleep disorders or other serious problems, such as heart disease or lung disease.
Portable EEG applications
Several studies have ventured into testing new technologies similar to portable EEG devices with smaller electrodes to perform sleep studies, obtaining very good results.
There have been numerous studies of portable monitors in the last two or three decades. The US government and medical societies have extensively reviewed this literature several times in an attempt to determine whether portable monitoring should be used more widely to diagnose disorders such as obstructive sleep apnea (OSA), for example. It is anticipated that portable monitoring as a diagnostic modality for OSA for example will be used more frequently. Physicians and others who consider the use of portable monitors should fully understand the advantages and limitations of this technology (Collop, 2008).
A couple of years ago, portable PSG was already being tested. Mykytyn, Sajkov, Neill and Mc Evoy (1999) compared the validity of the portable polysomnographic recorder with a laboratory-based polysomnographic system. The quality of the signals was comparable between the PSGP and PSGL studies. There was no significant discordance between PSGP and PSGL in the final diagnostic formulations, concluding that portable polysomnography is a viable alternative to laboratory-based polysomnography and can be further improved by a better sensor connection.
Fry, DiPhillipo, Curran, Goldberg and Baran (1998) evaluated the unattended complete PSG registered in the home and assessed the ability to acquire, store and analyze polysomnographic data. Using a digital recording system with miniature preamplifiers and the ability to record 18 channels of polysomnographic data, they found that full PSG without supervision can be performed at home with reliable and high quality recordings. The full PSG can be extended to a larger patient population, because it is no longer limited by the number of beds, and there is a reduction in costs due to the elimination of night staff and the cost of installation.
More recently, Rosen et al. (2012) conclude that a strategy in the home for diagnosis and treatment compared to PSG in the laboratory was not inferior in terms of acceptance, compliance, treatment time and functional improvements.
According to Lucey et al. (2016), a precise study on home sleep to assess sleep stages based on EEG and EEG potency would be advantageous both for clinical and research purposes, such as longitudinal studies that measure changes in sleep stages of the EEG over time. According to the authors, given that even limited PSG to monitor only sleep stages can be prohibitive for finances, unattended sleep monitoring that includes sleep stages and EEG potency but does not require placement of sensors by a technician would have an important clinical and scientific application for longitudinal research studies and clinical trials.
The staging of sleep and the measurement of spectral power with a single-channel EEG recorded on the forehead, as the Neurosky device has for example, presents multiple challenges compared to PSG that may affect the expected results. First and most obvious, the forehead canal will not detect an alpha rhythm and it is predicted that this will limit the evaluation of wakefulness and stage N1. Secondly, the waveforms used for sleep staging differ in the topographic distribution. The K complexes, for example, are distributed mainly over the mid-frontal branches, while the delta waves reach their maximum in both the frontal and occipital areas. In addition, the sleep spindles have a bimodal anterior-posterior distribution and the single EEG will allow the revision of the anterior sleep spindles. Also, age-related changes in spindle density, amplitude and duration are greater anteriorly. Thirdly, there are limitations on the ability of the single EEG to assess the topographic differences in slow wave activity (SWA) reported in adults or the subsequent change to previous SWA maximum activity reported during childhood. The NREM power also undergoes an anterior-posterior change during consecutive periods of NREM sleep that can not be assessed with a single forward lead. In addition, the measurement of SWA during REM may be less accurate than in a central derivation in a PSG due to eye movements (Lucey et al., 2016). However, the results of these authors show that the single-channel EEG provides results comparable to polysomnography in the evaluation of REM, combined sleep stages N2 and N3, and several other parameters, including the activity of slow forward wave. The data establishes that the single-channel EEG can be a useful research tool.
Sleep Staging and portable EEG: new data processing techniques
The staging of sleep involves the recording of the different stages of sleep and is part of the study of polysomniography, although a staging can also be carried out by means of an EEG device outside of a standard polysomniography, which usually involves the multiple additional records mentioned above (air flow in and out of the lungs during breathing, oxygen levels in the blood, body position, electrical activity of the muscles, heart rate, etc.). Staging allows us to have a first view of whether there may be problems at the level of sleep.
Several works have developed the possibility of automating the staging of sleep through various techniques. This suggests a great possibility to generate software applications that use portable EEG and that can be massively used remotely to evaluate basic aspects of sleep quality to allow the improvement of diagnosis and treatment, as well as generating applications that train people in hygiene of sleep.
The automatic classification of sleep is essential due to the fact that, conventionally, doctors have to visually analyze a large volume of data, which is onerous, time-consuming and prone to errors. Therefore, there is a great need for an automated sleep-staging scheme (Hassan & Bhuiyan, 2016). According to Principe, Gala and Chang (1989), the staging of sleep represents a problem of medical decision. The authors developed a model for automatic sleep staging by combining signal information, human heuristic knowledge in the form of rules and a mathematical framework. Liang, Kuo, Hu, Pan and Wang (2012) have proposed an automatic sleep scoring method that combines multiscale entropy (MEE) and autoregressive (AR) models for single-channel EEG to evaluate the performance of the method comparatively with the manual scoring in complete polysomnograms. Their results show that MEE is a useful and representative feature for sleep staging. It has high precision and good applicability for home care because only one EEG channel is used for sleep staging.
In addition, Liang, Kuo, Hu and Cheng (2012) developed an automatic rule-based sleep staging method with an average precision and a kappa coefficient that reached 86.68% and 0.79, respectively. The algorithm can be integrated with a portable PSG system for home sleep assessment. On the other hand, Koley and Dey (2012) worked on the automatic identification of various stages of sleep, such as sleep stages 1, 2, slow wave sleep (sleep stages 3 and 4), REM sleep and wakefulness for single-channel EEG signal. The automatic punctuation of the sleep stages was carried out with the help of a pattern recognition technique that involves the extraction of characteristics, selection and finally classification. The agreement of the estimated sleep stages with those obtained by experts for all sleep stages of the training data set was 0.877 and for the independent test data set it reached 0.8572. The proposed method could be used as an efficient and cost-effective way to stage sleep with the advantage of reducing stress and the burden imposed on the subjects.
On the other hand, Bajaj and Pachori (2013) applied a new method for the automatic classification of the sleep stage based on the time-frequency image (TFI) of the EEG signals. The results show good effectiveness of the proposed method for classifying the stages of sleep from EEG signals.
Also, Hassan and Bhuiyan (2016) proposed an automated sleep scoring method based on single channel EEG. A new technique of signal processing is used, the “TQWT” for the staging of sleep. The effectiveness of the method is confirmed by statistical analysis. The performance of the proposed scheme, compared to the existing ones, is promising. The proposed scheme will relieve the burden on physicians, accelerate the diagnosis of sleep disorder and accelerate sleep research.
Other advances have also been achieved regarding the nature of the EEG signals, which are are non-linear and non-stationary. It is difficult to perform sleep staging through visual interpretation and linear techniques (Acharya, Chua, Chua, Min, & Tamura, 2010). Therefore, the authors use a non-linear technique, the higher-order spectra (HOS), to extract information hidden in the EEG signal of the dream. Their results indicate that the proposed system is able to identify the stages of sleep with an accuracy of 88.7%.
Acharya, U. R., Chua, E. C. P., Chua, K. C., Min, L. C., & Tamura, T. (2010). Analysis and automatic identification of sleep stages using higher order spectra. International journal of neural systems, 20(06), 509-521.
Aguilera Olivares, L., Díaz, S., & Sánchez, G. (2012). Trastornos del sueño en el paciente adulto hospitalizado.
Ancoli-Israel, S. (1989). Epidemiology of sleep disorders. Clinics in geriatric medicine, 5(2), 347-362.
Bajaj, V., & Pachori, R. B. (2013). Automatic classification of sleep stages based on the time-frequency image of EEG signals. Computer methods and programs in biomedicine, 112(3), 320-328.
Benetó, A. (2000). Trastornos del sueño en el anciano. Epidemiología. Revista de Neurología, 30(6), 581-585.
Bixler, E. O., Kales, A., Soldatos, C. R., Kales, J. D., & Healey, S. (1979). Prevalence of sleep disorders in the Los Angeles metropolitan area. The American journal of psychiatry.
Collop, N. A. (2008). Portable monitoring for the diagnosis of obstructive sleep apnea. Current opinion in pulmonary medicine, 14(6), 525-529.
Convertini, G., Krupitzky, S., Tripodi, M. R., & Carusso, L. (2003). Trastornos del sueño en niños sanos. Arch argent pediatr, 101(2), 99-105.
Fry, J. M., DiPhillipo, M. A., Curran, K., Goldberg, R., & Baran, A. S. (1998). Full polysomnography in the home. Sleep, 21(6), 635-642.
Fursow, Y. B., Betancourt, C. J., & León, J. C. J. (2006). Trastorno por déficit de atención e hiperactividad y trastornos del sueño. Revista de neurología, 42(2), 37-51.
Hassan, A. R., & Bhuiyan, M. I. H. (2016). A decision support system for automatic sleep staging from EEG signals using tunable Q-factor wavelet transform and spectral features. Journal of neuroscience methods, 271, 107-118.
Hillman, D. R., Murphy, A. S., Antic, R., & Pezzullo, L. (2006). The economic cost of sleep disorders. Sleep, 29(3), 299-305.
Jiménez, M. A. G., Redondo-Martínez, M. P., Marcos-Navarro, A. I., Torrijos-Martínez, M. P., Aguilar, F. S., Monterde-Aznar, M. L., & Rodríguez-Almonacid, F. M. (2004). Prevalencia de los trastornos del sueño en adolescentes de Cuenca, España. Revista de neurología, 39(1), 18-24.
Koley, B., & Dey, D. (2012). An ensemble system for automatic sleep stage classification using single channel EEG signal. Computers in biology and medicine, 42(12), 1186-1195.
Liang, S. F., Kuo, C. E., Hu, Y. H., & Cheng, Y. S. (2012). A rule-based automatic sleep staging method. Journal of neuroscience methods, 205(1), 169-176.
Liang, S. F., Kuo, C. E., Hu, Y. H., Pan, Y. H., & Wang, Y. H. (2012). Automatic Stage Scoring of Single-Channel Sleep EEG by Using Multiscale Entropy and Autoregressive Models. IEEE Trans. Instrumentation and Measurement, 61(6), 1649-1657.
Lucey, B. P., Mcleland, J. S., Toedebusch, C. D., Boyd, J., Morris, J. C., Landsness, E. C., … & Holtzman, D. M. (2016). Comparison of a single‐channel EEG sleep study to polysomnography. Journal of sleep research, 25(6), 625-635.
Mykytyn, I. J., Sajkov, D., Neill, A. M., & Mc Evoy, R. D. (1999). Portable computerized polysomnography in attended and unattended settings. Chest, 115(1), 114-122.
Owens, J. (2007). Classification and epidemiology of childhood sleep disorders. Sleep Medicine Clinics, 2(3), 353-361
Principe, J. C., Gala, S. K., & Chang, T. G. (1989). Sleep staging automaton based on the theory of evidence. IEEE Transactions on Biomedical Engineering, 36(5), 503-509.
Rosen, C. L., Auckley, D., Benca, R., Foldvary-Schaefer, N., Iber, C., Kapur, V., … & Redline, S. (2012). A multisite randomized trial of portable sleep studies and positive airway pressure autotitration versus laboratory-based polysomnography for the diagnosis and treatment of obstructive sleep apnea: the HomePAP study. Sleep, 35(6), 757-767.
Sateia, M. J. (2014). International classification of sleep disorders. Chest, 146(5), 1387-1394.
Sierra, J. C., Jiménez-Navarro, C., & Martín-Ortiz, J. D. (2002). Calidad del sueño en estudiantes universitarios: importancia de la higiene del sueño. Salud mental, 25(6), 35-43.
Our work at Naxon for professionals aligns with global standards that involve promoting well-being and human development. The advantages of the technology of portable EEG devices applied to the area of Medicine and Clinical Psychology are substantial and significantly improve several current problems, as will be described below. The rapid advances in the use of software technology for the analysis of patterns and the conformation of information processing algorithms allow us to work with these devices generating automatic processing of information that reduces the workload of professionals, increases their efficiency and improves the diagnoses and consequent treatments.
Following authors Mirza et al. 2015, improving the quality of life of the elderly and disabled and providing them with adequate care at the right time is one of the most important roles we must play as responsible members of society. Before a few decades, people with disabilities could not perform daily tasks such as turning on the light, making a phone call and even controlling the TV. Disability was an obstacle, so people with impairments needed supervision and daily services. With the evolution of computing power and progress in neural studies, neurotechnology and cybernetics, modern technology and science is managing to overcome human disability (Rihana, Azar, & Bitar, 2016).
A recent development within the area of wearable technology and brain-computer interfaces (BCI) has been the creation of portable EEG devices that connect to a computer, tablet or smartphone via Bluetooth, which can really expand the development of applications given this connectivity, as well as the low cost they present, added to the non-invasive qualities of the devices. These headsets use a biosensor that allows capturing brainwaves which can be used to develop an app for mind-controlled hardware as well as other applications such as Neurofeedback training therapy, meditation, managing sleep, assessing emotions, among others.
Modern BCI applications have proven to be efficient in “thought controlled devices”. A typical example is found in a wheelchair controlled by thought, which uses captured signals from the brain and eyes and processes them to control its movements. The EEG technique deploys an electrode cap that is placed on the user’s scalp for the acquisition of signals that are captured and translated into movement commands that in turn move the wheelchair (Mirza et al., 2015). Many EEG commands can be used. For example, the low-cost portable system has been designed and tested to allow a person to control the television through blinking with high precision. A prototype did not exceed 60 usd (Rihana, Azar, & Bitar, 2016).
In another context, in the intensive care unit, or during anesthesia, patients are connected to monitors by means of cables. These cables obstruct the nursing staff and prevent patients from moving freely in the hospital. However, it is expected that rapidly developing wireless technologies will solve these problems, one of which is portable EEG (Paksuniemi, Sorvoja, Alasaarela, & Myllyla, 2006).
The ability of telemonitoring in relation to software applications is a key feature of these devices, and is defined as the use of information technology to monitor patients remotely. The review of the literature suggests that the most promising applications for telemonitoring are chronic diseases. The monitoring allows to reduce the complications of chronic diseases thanks to a better follow-up; provides health care services without using hospital beds; and it reduces the patient’s travel, time away from work and general costs. Several systems have proven to be profitable. Telmonitoring is also a way to respond to the new needs of home care in an aging population. The obstacles to the development of telemonitoring include the initial costs of the systems, the doctors’ licenses and the reimbursement (Meystre, 2005). A key feature of portable EEGs are their relatively low costs, even compared to the simpler new hardware used by many clinics today.
Recently, the versatility of this EEG technology has been tested, managing complicated tasks with only mind control, such as playing a tactical video game. In this particular case, reliable control is achieved by adapting a steady state evoked visual potential classifier (SSVEP), a type of measurement obtained with EEG, to be robust enough to cope with signal quality of the portable EEG device used, in this case the Emotiv Epoc. The difference in the performance of the game that runs on more advanced EEG equipment of the “research” type compared to the EPOC Emotiv has been examined, having a satisfactory response from the public (van Vliet et al., 2012).
Conventional brain-computer interfaces are often expensive, complex to operate, and lack portability, which limits their use in laboratory settings for example. Portable and cheap BCIs can mitigate these problems, but their performance must be proven. Several authors have conducted promising research in this regard. McCrimmon et al. (2017) developed a low cost portable BCI with performance comparable to conventional BCIs. These platforms are suitable for BCI applications outside of a laboratory. According to Krigolson, Williams, Norton, Hassall and Colino (2017), the validation of the use of low cost EEG systems has focused on the continuous recording of EEG data and/or the replication of EEG configurations of large systems that depend on event markers to allow the examination of brain events related potentials (ERP). These authors demonstrate that it is possible to perform an ERP research without relying on event markers using a portable Muse EEG system and a single computer. His work emphasizes that with a single computer and a portable EEG system such as Muse, ERP research can be easily conducted, thus extending the possible use of the methodology to a variety of novel contexts. At the same time, it shows the promising capabilities of these devices. In turn, Armanfard, Komeili, Reilly and Pino (2016) proposed a practical approach to machine learning for the identification of mental vigilance failures using portable EEG signals recorded from a very sparse electrode configuration with only 4 channels. This is a difficult problem since these four electrodes are easily contaminated with flickers and muscle artifacts. The performance of the algorithm based on the proposed machine learning is demonstrated in a real world scenario in which the surveillance lapses are identified with approximately 95% accuracy, adding greater validation to these eeg headsets.
On the other hand, the scientific community is working with several tools for the processing and classification of signals, which helps to develop additional applications. Karydis, Aguiar, Foster and Mershin (2015) propose a new autocalibration protocol that can solidly classify brain states when combined with five standard machine learning algorithms. Their results indicate that the portable EEG sensors available in the market provide sufficient data fidelity at sufficiently fast speeds to differentiate in real time the arbitrary states defined by the user. The classification of EEG signals is an important task in BCI. Zhang, Liu, Ji and Huang (2017) present two combined strategies of feature extraction in EEG signals using autoregressive coefficients and approximate EEG entropy of decay signals in subbands, which has also very satisfactory results.
An important aspect in the non-invasive research of the brain-computer interface is to acquire the EGG in an adequate manner. From the perspective of the final user, this means with maximum comfort and without any additional inconvenience (for example, washing the hair), while from a technical perspective, the quality of the signal must be optimal for the BCI to function effectively and efficiently (Pinegger, Wriessnegger, Faller, & Müller-Putz, 2016). These authors found that the construction of a reliable BCI is possible with all the systems assessed, and it is the responsibility of the user to decide which system best meets the given requirements.
Thus, the brain-computer interface has been investigated to a greater extent in the area of robotics and medicine, and not so much in direct applications to the consumer. As processors have advanced in terms of a surprisingly high performance/cost ratio, BCIs for popular use outside the medical environment has taken great strides to reach consumers in the last decade (Chu, 2017). In this sense McKenzie et al. (2017) evaluated the ability of an smartphone based EGG application, the Smartphone Brain Scanner-2 (SBS2), to detect epileptiform abnormalities compared to the standard clinical EEG. Epileptiform discharges were detected in 14% of SBS2 and in 25% of standard EEG. Despite limitations in sensitivity, SBS2 can become a viable support test for the capture of epileptiform anomalies and extend access to EEG to new populations, especially of limited resources, at a reduced cost. All this research shows the potential to develop cheap software applications that can be used with low-cost personal devices that can be used massively by the population, expanding the accessibility of diagnoses and treatments.
Armanfard, N., Komeili, M., Reilly, J. P., & Pino, L. (2016, May). Vigilance lapse identification using sparse EEG electrode arrays. In Electrical and Computer Engineering (CCECE), 2016 IEEE Canadian Conference on (pp. 1-4). IEEE.
Chu, N. N. (2017). Surprising Prevalence of Electroencephalogram Brain-Computer Interface to Internet of Things [Future Directions]. IEEE Consumer Electronics Magazine, 6(2), 31-39.
McCrimmon, C. M., Fu, J. L., Wang, M., Lopes, L. S., Wang, P. T., Karimi-Bidhendi, A., … & Do, A. H. (2017). Performance Assessment of a Custom, Portable, and Low-Cost Brain–Computer Interface Platform. IEEE Transactions on Biomedical Engineering, 64(10), 2313-2320.
McKenzie, E. D., Lim, A. S., Leung, E. C., Cole, A. J., Lam, A. D., Eloyan, A., … & Bui, E. (2017). Validation of a smartphone-based EEG among people with epilepsy: A prospective study. Scientific reports, 7, 45567.
Meystre, S. (2005). The current state of telemonitoring: a comment on the literature. Telemedicine Journal & e-Health, 11(1), 63-69.
Mirza, I. A., Tripathy, A., Chopra, S., D’Sa, M., Rajagopalan, K., D’Souza, A., & Sharma, N. (2015, February). Mind-controlled wheelchair using an EEG headset and arduino microcontroller. In Technologies for Sustainable Development (ICTSD), 2015 International Conference on (pp. 1-5). IEEE.
Karydis, T., Aguiar, F., Foster, S. L., & Mershin, A. (2015, July). Self-calibrating protocols enhance wearable EEG diagnostics and consumer applications. In Proceedings of the 8th ACM International Conference on Pervasive Technologies Related to Assistive Environments (p. 96). ACM.
Krigolson, O. E., Williams, C. C., Norton, A., Hassall, C. D., & Colino, F. L. (2017). Choosing MUSE: Validation of a low-cost, portable EEG system for ERP research. Frontiers in neuroscience, 11, 109.
Rihana, S., Azar, T., & Bitar, E. (2016, October). Portable EEG recording system for BCI application. In Biomedical Engineering (MECBME), 2016 3rd Middle East Conference on (pp. 80-83). IEEE.
Paksuniemi, M., Sorvoja, H., Alasaarela, E., & Myllyla, R. (2006, January). Wireless sensor and data transmission needs and technologies for patient monitoring in the operating room and intensive care unit. In Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual International Conference of the (pp. 5182-5185). IEEE.
Pinegger, A., Wriessnegger, S. C., Faller, J., & Müller-Putz, G. R. (2016). Evaluation of different EEG acquisition systems concerning their suitability for building a brain–computer interface: case studies. Frontiers in neuroscience, 10, 441.
van Vliet, M., Robben, A., Chumerin, N., Manyakov, N. V., Combaz, A., & Van Hulle, M. M. (2012, January). Designing a brain-computer interface controlled video-game using consumer grade EEG hardware. In Biosignals and Biorobotics Conference (BRC), 2012 ISSNIP (pp. 1-6). IEEE.
Zhang, Y., Liu, B., Ji, X., & Huang, D. (2017). Classification of EEG signals based on autoregressive model and wavelet packet decomposition. Neural Processing Letters, 45(2), 365-378.
There is an increasing innovation in the clinical psychological field and in the area of personal wellbeing in general, driven by the incorporation of important technological advances that have a significant impact in various disciplines within health.
A recent case for example, has been the considerable progress in the incorporation of technologies such as virtual reality for the psychological treatment and management of phobias and situations of stress and anguish, to such an extent that it has become an important ally of professionals to improve the quality of life of patients.
For those who are not aware of another recent development, Brain-Computer Interfaces or BCIs are having a similar impact, although we are just beginning to explore their applicability for professionals in health.
A BCI represents a direct communication path between an improved or connected brain and an external device. It can be invasive in the form of electrodes implanted directly in the brain or non-invasive through an electroencephalogram or EEG, for example, which reads the electrical brain waves in our scalp that result in commands executed by an external hardware and software.
A recent development in BCI has been the creation of portable EEG devices that connect to a computer, tablet or smartphone through Bluetooth, which can really expand the development of applications given this connectivity, as well as the low cost they present, added to the non-invasive qualities. Among these devices is Muse developed by the Canadian company Interaxon. It is marketed as a device for anyone who wants to train relaxation and mindfulness. It is also marketed for mental health professionals to work on these aspects linked to the relaxation response and mindfulness.
By reading the brain waves produced by the brain, the Muse can detect if the levels of attention and relaxation increase or decrease during a meditation session. Through a system of feedback based on different environmental or musical sounds, the device translates neuronal information into useful signals for the person: if the ambient sound increases in intensity the person is less relaxed and focused and vice versa.
As part of my work at Naxon, where we are developing practical applications with portable EEG devices, I had the opportunity to test for a while the use of the original Muse and Muse 2 to train meditation.
One of the points to highlight given my personal experience is that the use of the device helps a lot to encourage the daily practice of meditation and promotes a positive feedback to get involved with this healthy practice.
In turn, the Muse provides a detailed record of your meditation activity that allows one to keep track of the evolution of the practice and how the training impacts over time in our anxious responses and in the speed and effectiveness by which we managed to induce a relaxation response. In general, the experience has been very comforting and I believe that it has a very good potential in clinical practice for health professionals, for the reasons described above, since it allows the professional to have an accurate record of certain cognitive-emotional responses related to stress and relaxation, and at the same time it represents a tool to work on different techniques in sessions with clients to control anxiety and stress. In turn, unlike a more traditional approach, the use of a novel technological tool allows the client to be more involved and also to monitor a practice remotely that the client can perform at home with the Muse device of the therapist.
This record is also very useful for having an accurate measurement of the particular training of a person in meditation, a measure that is very important when conducting scientific studies involving novice individuals or experts in practice.
The first model of the device uses brain wave reading to work on meditation. The renewed Muse 2 also offers a measure of heart frequency, the activity of our breathing and the movements of our body. With these new signals, new types of meditations based on these measures can be performed, which make excellent anxiety and stress response markers, which provide even more assistance for professionals, as well as for users who wish to start the practice of meditation or want to improve their experience with it.
Yesterday we met with our ANII project executive to discuss the details related to the funding we have obtained from the agency. ANII is the most important institution in Uruguay in terms of research funding and support of innovative projects and startups in the academic-scientific, technological and entrepreneurial fields. We are very happy and grateful that our proposal has been received with enthusiasm.
During the end of last year and the beginning of 2019 we had been working on the application for funding at this agency within the line “Tools for Innovation”. On May 21st, the ANII evaluation committee approved our project, with a very positive feedback regarding the interest generated and the prospects for future work.
Based on our previous work and the private investment with which the initiative already counted, we have been able to make considerable progress in the products we are developing. This new impulse by ANII implies increasing advances and a promising future.
Yesterday we met with the guys of Eagerworks to continue working with our developments.
A productive session where we celebrate the advances. At the same time, challenges arise which encourage us to work hard to make the best product for professionals, with new ideas to incorporate so as to generate new and innovative services that will benefit our future clients.
In July 2017, under the code name “Naxon“, exchanging ideas from the neuroscientific and information technology perspectives, the initiative to work with portable EEG technologies emerged with the aim of developing practical and economic solutions for some of the problems present in the areas of Health and Well-being. Maturing the idea during 2018 we began to experiment with Neurosky and Muse devices, and the first stage of developmentof the venture was conceptualized.
A Brain-Computer interface or BCI is a direct communication pathway between an enhanced or wired brain and an external device. These can be invasive in the form of electrodes implanted directly on the brain or non-invasive through an EEG for example, that reads the electrical brain waves on our scalp that are translated into commands that an external hardware and software executes.
You can see the following video for an introduction on the subject: link
As shown in the video, BCI can be used for a variety of things, such as helping people with physical disabilities to adapt better through the use of a robotic hand controlled by the mind for example, or use BCI in gaming to execute different commands, or to control other external hardware, like a drone (see the following video). Apart from these examples, the technology allows for many applications in different fields the have not yet been explored.
One recent development in BCI has been the creation of portable EEG devices that connect to a computer, tablet or smartphone through Bluetooth such as the one seen in the previous video, that can really expand the development of applications given this connectivity, but also the low cost and non-invasiveness qualities of the device.
One of the basic elements of brain functioning lies in the synapse. This is the point where electrical signals from one neuron generate electrical effects in another. The initiation of an action potential, for example, that which allows transmitting nerve impulses, and which is the basis of brain functioning at the level of the neuron, requires the opening of sodium ion channels in the mound of the axon of the neuron. At this level, the phenomena depend on physical and chemical properties and on the characteristics of the biological structures at play. To unleash an action potential, a certain trigger threshold must be overcome thanks to the action of graduated potentials that are added temporally and spatially, provided by the connections with other neurons. In order for activity to occur in the latter, the same phenomena must occur (Kandel, Schwartz, & Jessel, 2001). These EEG devices exploit this electrical property of the brain. Neurons signal electrically. The potential (voltage) across their cell membranes changes, when channels in the membrane open and ions move down their electrochemical gradients.
There are different ways of measuring brain activity, all can be classified according to temporal resolution, spatial resolution and invasiveness:
The classical electrophysiological monitoring method records electrical activity of the brain, which is mainly noninvasive, with electrodes placed along the scalp, although invasive electrodes are sometimes used such as in electrocorticography. Associated to the use of EEG, we have event-related potentials which investigate potential fluctuations time locked to an event like stimulus onset or button press. Also, spectral content of EEG: analyses the type of neural oscillations (popularly called “brain waves”) that can be observed in EEG signals in the frequency domain.
EEG reflects large-scale neural activity, producing tiny electrical dipoles which transduce through the skull. However, thiselectrical activity is incredibly small and must be amplified thousands of times.
Besides research use, EEG is usually used in clinical contextsfordiagnostic applicationssuch as in the case of epilepsy, which causes abnormalities in EEG readings.Also in sleep disorders, depth of anesthesia, coma, encephalopathies, and brain death. It is also a first-line method of diagnosis for tumors, stroke and other focal brain disorders, but this use has decreased with the advent of high-resolution anatomical imaging techniques such as magnetic resonance imaging (MRI) and computed tomography (CT).
Despite limited spatial resolution, EEG continues to be a valuable tool for research and diagnosis. It is one of the few mobile techniques available and offers millisecond-range temporal resolutionwhich is not possible with CT, PET or MRI.Derivatives of the EEG technique likeevoked potentials (EP), are useful, which involves averaging the EEG activity time-locked to the presentation of a stimulus of some sort (visual, somatosensory, or auditory).Event-related potentials (ERPs) on the other hand,refer to averaged EEG responses that are time-locked to more complex processing of stimuli; this technique is used in cognitive science, cognitive psychology, and psychophysiological research.ERPs can be collected across the lifespan. This has implications for obtaining early and objective rather than late and subjective measures of sensory and cognitive sensitivity.Participants do not have to be actively listening to record meaningful auditory ERP for example. This has implications for assessing neural activity in the resting brain, coma and other such states.Auditory (and visual) ERPs can also be assessed under varying conditions of task load and intention. This might help us uncover memory and attention processes.The new portable devices typically have fewer electrodes, but have the advantage of rapid setting, without the need to use gels and be on a lab, are much cheaper, they can be used in different settings and connect wirelessly to a computer device for example.
Some of these devices, like we previously mentioned are the Emotiv, Neurosky and Muse. The Muse device for example is sold as a meditation tool together with a mobile application that gives one auditory feedback of one’s state of mind, giving us indications of when we are calm or mind wandering and more anxious.