Publications by Dr Palaniappan Ramaswamy

Perturbations in the autonomic nervous system occur in individuals experiencing increasing levels of motion sickness. Here, we investigated the effects of transauricular electrical stimulation (tES) on autonomic function during visually induced motion sickness, through the analysis of spectral and time-frequency heart rate variability. To determine the efficacy of tES, we compared sham and tES conditions in a randomized, within-subjects, cross-over design in 14 healthy participants. We found that tES reduced motion sickness symptoms by significantly increasing normalized high-frequency (HF) power and decreasing both normalized low-frequency (LF) power and the power ratio of LF and HF components (LF/HF ratio). Furthermore, behavioral data recorded using the motion sickness assessment questionnaire (MSAQ) showed significant differences in decreased symptoms during tES compared to sham condition for the total MSAQ scores and, central and sopite categories of the MSAQ. Our preliminary findings suggest that by administering tES, parasympathetic modulation is increased, and autonomic imbalance induced by motion sickness is restored. This study provides first evidence that tES may have potential as a non-pharmacological neuromodulation tool to keep motion sickness at bay. Thus, these findings may have implications towards protecting people from becoming motion sick and possible accelerated recovery from the malady.

Hand gesture recognition with surface electromyography (sEMG) is indispensable for Muscle-Gesture-Computer Interface. The usual focus of it is upon performance evaluation involving the accuracy and robustness of hand gesture recognition. However, addressing the reliability of such classifiers has been absent, to our best knowledge. This may be due to the lack of consensus on the definition of model reliability in this field. An uncertainty-aware model has the potential to self-evaluate the quality of its inference, thereby making it more reliable. Moreover, uncertainty-based rejection has been shown to improve the performance of sEMG-based hand gesture recognition. Therefore, we first define model reliability here as the quality of its uncertainty estimation and propose an offline framework to quantify it. To promote reliability analysis, we propose a novel end-to-end uncertainty-aware finger movement classifier, i.e., evidential convolutional neural network (ECNN), and illustrate the advantages of its multidimensional uncertainties such as vacuity and dissonance. Extensive comparisons of accuracy and reliability are conducted on NinaPro Database 5, exercise A, across CNN and three variants of ECNN based on different training strategies. The results of classifying 12 finger movements over 10 subjects show that the best mean accuracy achieved by ECNN is 76.34%, which is slightly higher than the state-of-the-art performance. Furthermore, ECNN variants are more reliable than CNN in general, where the highest improvement of reliability of 19.33% is observed. This work demonstrates the potential of ECNN and recommends using the proposed reliability analysis as a supplementary measure for studying sEMG-based hand gesture recognition.

Half of all road accidents result from either lack of driver attention or from maintaining insufficient separation between vehicles. Collision from the rear, in particular, has been identified as the most common class of accident in the UK, and its influencing factors have been widely studied for many years. Rear-mounted stop lamps, illuminated when braking, are the primary mechanism to alert following drivers to the need to reduce speed or brake. This paper develops a novel brain response approach to measuring subject reaction to different brake light designs. A variety of off-the-shelf brake light assemblies are tested in a physical simulated driving environment to assess the cognitive reaction times of 22 subjects. Eight pairs of LED-based and two pairs of incandescent bulb-based brake light assemblies are used and electroencephalogram (EEG) data recorded. Channel Pz is utilised to extract the P3 component evoked during the decision making process that occurs in the brain when a participant decides to lift their foot from the accelerator and depress the brake. EEG analysis shows that both incandescent bulb-based lights are statistically slower to evoke cognitive responses than all tested LED-based lights. Between the LED designs, differences are evident, but not statistically significant, attributed to the significant amount of movement artifact in the EEG signal

This paper presents and explores a robust deep learning framework for auscultation analysis. This aims to classify anomalies in respiratory cycles and detect diseases, from respiratory sound recordings. The framework begins with front-end feature extraction that transforms input sound into a spectrogram representation. Then, a back-end deep learning network is used to classify the spectrogram features into categories of respiratory anomaly cycles or diseases. Experiments, conducted over the ICBHI benchmark dataset of respiratory sounds, confirm three main contributions towards respiratory- sound analysis. Firstly, we carry out an extensive exploration of the effect of spectrogram types, spectral-time resolution, overlapping/non-overlapping windows, and data augmentation on final prediction accuracy. This leads us to propose a novel deep learning system, built on the proposed framework, which outperforms current state-of-the-art methods. Finally, we apply a Teacher-Student scheme to achieve a trade-off between model performance and model complexity which holds promise for building real-time applications.

Rear-end collision accounts for around 8% of all vehicle crashes in the UK, with the failure to notice or react to a brake light signal being a major contributory cause. Meanwhile traditional incandescent brake light bulbs on vehicles are increasingly being replaced by a profusion of designs featuring LEDs. In this paper, we investigate the efficacy of brake light design using a novel approach to recording subject reaction times in a simulation setting using physical brake light assemblies. The reaction times of 22 subjects were measured for ten pairs of LED and incandescent bulb brake lights. Three events were investigated for each subject, namely the latency of brake light activation to accelerator release (BrakeAcc), the latency of accelerator release to brake pedal depression (AccPdl), and the cumulative time from light activation to brake pedal depression (BrakePdl). To our knowledge, this is the first study in which reaction times have been split into BrakeAcc and AccPdl. Results indicate that the two brake lights containing incandescent bulbs led to significantly slower reaction times compared to eight tested LED lights. BrakeAcc results also show that experienced subjects were quicker to respond to the activation of brake lights by releasing the accelerator pedal. Interestingly, analysis also revealed that the type of brake light influenced the AccPdl time, although experienced subjects did not always act quicker than inexperienced subjects. Overall, the study found that different designs of brake light can significantly influence driver response times.

This article proposes an encoder-decoder network model for Acoustic Scene Classification (ASC), the task of identifying the scene of an audio recording from its acoustic signature. We make use of multiple low-level spectrogram features at the front-end, transformed into higher level features through a well-trained CNN-DNN front-end encoder. The high-level features and their combination (via a trained feature combiner) are then fed into different decoder models comprising random forest regression, DNNs and a mixture of experts, for back-end classification. We conduct extensive experiments to evaluate the performance of this framework on various ASC datasets, including LITIS Rouen and IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE) 2016 Task 1, 2017 Task 1, 2018 Tasks 1A & 1B and 2019 Tasks 1A & 1B. The experimental results highlight two main contributions; the first is an effective method for high-level feature extraction from multi-spectrogram input via the novel CNN-DNN architecture encoder network, and the second is the proposed decoder which enables the framework to achieve competitive results on various datasets. The fact that a single framework is highly competitive for several different challenges is an indicator of its robustness for performing general ASC tasks.

Steady State Visual Evoked Potential (SSVEP) methods for brain–computer interfaces (BCI) are popular due to higher information transfer rate and easier setup with minimal training, compared to alternative methods. With precisely generated visual stimulus frequency, it is possible to translate brain signals into external actions or signals. Traditionally, SSVEP data is collected from the occipital region using electrodes with or without gel, normally mounted on a head cap. In this experimental study, we develop an in-ear electrode to collect SSVEP data for four different flicker frequencies and compare against occipital scalp electrode data. Data from five participants demonstrates the feasibility of in-ear electrode based SSVEP, significantly enhancing the practicability of wearable BCI applications.

A fully customisable chip-on board (COB) LED design to evoke two brain responses simultaneously (steady state visual evoked potential (SSVEP) and transient evoked potential, P300) is discussed in this paper. Considering different possible modalities in brain-computer interfacing (BCI), SSVEP is widely accepted as it requires a lesser number of electroencephalogram (EEG) electrodes and minimal training time. The aim of this work was to produce a hybrid BCI hardware platform to evoke SSVEP and P300 precisely with reduced fatigue and improved classification performance. The system comprises of four independent radial green visual stimuli controlled individually by a 32-bit microcontroller platform to evoke SSVEP and four red LEDs flashing at random intervals to generate P300 events. The system can also record the P300 event timestamps that can be used in classification, to improve the accuracy and reliability. The hybrid stimulus was tested for real-time classification accuracy by controlling a LEGO robot to move in four directions.

This paper explores the use of three different two-dimensional time-frequency features for audio event classification with deep neural network back-end classifiers. The evaluations use spectrogram, cochleogram and constant-Q transform based images for classification of 50 classes of audio events in varying levels of acoustic background noise, revealing interesting performance patterns with respect to noise level, feature image type and classifier. Evidence is obtained that two well-performing features, the spectrogram and cochleogram, make use of information that is potentially complementary in the input features. Feature fusion is thus explored for each pair of features, as well as for all tested features. Results indicate that a fusion of spectrogram and cochleogram information is particularly beneficial, yielding an impressive 50-class accuracy of over 96% in 0dB SNR, and exceeding 99% accuracy in 10dB SNR and above. Meanwhile the cochleogram image feature is found to perform well in extreme noise cases of -5dB and -10dB SNR.

The students' progression and attainment gap are considered as key performance indicators of many universities worldwide. Therefore, universities invest significantly in resources to reduce the attainment gap between good and poor performing students. In this regard, various mathematical models have been utilised to predict students' performances in the hope of informing the support team to intervene at an early stage of the at risk student's at the university. In this work, we used a combination of institutional, academic, demographic, psychological and economic factors to predict students' performances using a multi-layered neural network (NN) to classify students' degrees into either a good or basic degree class. To our knowledge, the usage of such an extended profile is novel. A feed-forward network with 100 nodes in the hidden layer trained using Levenberg-Marquardt learning algorithm was able to achieve the best performance with an average classification accuracy of 83.7%, sensitivity of 77.37%, specificity of 85.16%, Positive Predictive Value of 94.04%, and Negative Predictive Value of 50.93%. The NN model was also compared against other classi?ers specifically k-Nearest Neighbour, Decision Tree and Support Vector Machine on the same dataset using the same features. The results indicate that the NN outperforms all other classi?ers in terms of overall classification accuracy and shows promise for the method to be used in Student Success ventures in the universities in an automatic manner.

Currently, carbon dioxide (CO2) waveforms measured by capnography are used to estimate respiratory rate and end-tidal CO2 (EtCO2) in the clinic. However, the shape of the CO2 signal carries significant diagnostic information about the asthmatic condition. Previous studies have shown a strong correlation between various features that quantitatively characterize the shape of CO2 signal and are used to discriminate asthma from non-asthma using pulmonary function tests, but no reliable progress was made, and no translation into clinical practice has been achieved. Therefore, this study reports a relatively simple signal processing algorithm for automatic differentiation of asthma and non-asthma. CO2 signals were recorded from 30 non-asthmatic and 43 asthmatic patients. Each breath cycle was decomposed into subcycles, and features were computationally extracted. Thereafter, feature selection was performed using the area (Az) under the receiver operating characteristics curve analysis. A classification was performed via a leave-one-out (LOO) cross-validation procedure by employing a support vector machine (SVM). Our results show maximum screening capabilities for upward expiration (AR1), downward inspiration (AR2) and the sum of AR1 and AR2, with an Az of 0.892, 0.803, and 0.793, respectively. The proposed method obtained an average accuracy of 94.52%, sensitivity of 97.67%, and specificity of 90% for discrimination of asthma and non-asthma. The proposed method allows for automatic classification of asthma and non-asthma condition by analyzing the shape of the CO2 waveform. The developed method may possibly be incorporated in real-time for assessment and management of the asthmatic conditions.

Steady state visual evoked response (SSVEP) is widely used in visual based diagnosis and applications such as brain-computer interfacing due to its high information transfer rate and the capability to activate commands through simple gaze control. However, one major impediment in using flashing visual stimulus to obtain SSVEP is eye fatigue that prevents continued long term use preventing practical deployment. This combined with the difficulty in establishing precise pulse width modulation (PWM) that results in poorer accuracy warrants the development of appropriate approach to solve these issues. Various studies have suggested the usage of high frequencies of visual stimulus to reduce the visual fatigue for the user but this results in poor response performance. Here, we study the use of extremely high duty cycles in the stimulus in the hope of solving these constraints. Electroencephalogram (EEG) data was recorded with PWM duty cycles of 50-95% generated by a precise custom made LED hardware and tested ten subjects responded that increasing duty cycles had less visual strain for all the frequency values and the SSVEP exhibited a subject independent peak response for duty cycle of 85%. This could pave the way for increased usage of SSVEP for practical applications.

Binaural beats are said to occur when two tones of different frequencies are presented to each ear, with the resulting perception being that of a single tone and a frequency midway between the two stimuli. It has found use in applications that purport to have the ability to relax the brain and other positive medical outcomes. This research considers the effects of binaural beats on the brain, following the assumption that the human brain has the tendency to change its dominant electroencephalography (EEG) frequency towards the frequency of the external stimulus. First, we considered the difference in spectral power of EEG frequencies for a group of individuals when in an entrained state (exposed to binaural beats) and a relaxed state (not exposed to binaural beats). Secondly, we assessed whether the effects of binaural beats affect the right and left hemisphere of the brain uniformly, and if so, to what extent this interhemispheric relationship can be seen. The results of this research show that hemispheric synchronisation affects all 5 subjects. Spectral power analysis revealed that 3 of the 5 subjects exhibited a difference in spectral power in the entrained state versus the relaxed state, especially in the frontal cortex but rather, in contradiction to the reports by others, as the effect was opposite, that is, entrainment reduces the spectral power, instead of causing an increase. The conclusion (based on the limited pool of subjects) is that there is some evidence to support that binaural beats do affect the brain but, in certain cases, negatively, and therefore, its usage should be weighed carefully.

The objective of this research was to investigate the relationship between emotion recognition and lateralization of motor onset in Parkinson's disease (PD) patients using electroencephalogram (EEG) signals. The subject pool consisted of twenty PD patients [ten with predominantly leftsided (LPD) and ten with predominantly rightsided (RPD) motor symptoms] and 20 healthy controls (HC) that were matched for age and gender. Multimodal stimuli were used to evoke simple emotions, such as happiness, sadness, fear, anger, surprise, and disgust. Artifactfree emotion EEG signals were processed using the auto regressive spectral method and then subjected to repeated ANOVA measures. No group differences were observed across behavioral measures? however, a significant reduction in EEG spectral power was observed at alpha, beta and gamma frequency oscillations in LPD, compared to RPD and HC participants, suggesting that LPD patients (inferred righthemisphere pathology) are impaired compared to RPD patients in emotional processing. We also found that PD related emotional processing deficits may be selective to the perception of negative emotions. Previous findings have suggested a hemispheric effect on emotion processing that could be related to emotional response impairment in a subgroup of PD patients. This study may help in clinical practice to uncover potential neurophysiologic abnormalities of emotional changes with respect to PD patient's motor onset.

Multiresolution analysis (MRA) over graph representation of EEG data has proved to be a promising method for offline brain–computer interfacing (BCI) data analysis. For the first time we aim to prove the feasibility of the graph lifting transform in an online BCI system. Instead of developing a pointer device or a wheel-chair controller as test bed for human–machine interaction, we have designed and developed an engaging game which can be controlled by means of imaginary limb movements. Some modifications to the existing MRA analysis over graphs for BCI have also been proposed, such as the use of common spatial patterns for feature extraction at the different levels of decomposition, and sequential floating forward search as a best basis selection technique. In the online game experiment we obtained for three classes an average classification rate of 63.0% for fourteen naive subjects. The application of a best basis selection method helps significantly decrease the computing resources needed. The present study allows us to further understand and assess the benefits of the use of tailored wavelet analysis for processing motor imagery data and contributes to the further development of BCI for gaming purposes.

This article focuses on different orientations of LED visual stimulus configurations which could be used to elicit Steady State Visual Evoked Potentials (SSVEP). SSVEP is extensively used in the research for various biomedical applications and require a configurable light source flickering at a constant frequency that would induce responses in corresponding frequencies in the EEG recorded over the visual cortex area of the scalp. The present study investigated the SSVEP amplitude dependence of horizontal and vertical LED visual stimulus orientations. SSVEP amplitudes were compared for five healthy subjects from five sessions of 30 seconds each and the recorded EEG signals were analysed. Individual recording sessions were carried out with horizontal and vertical orientation with 10 Hz visual stimulus for analysing the SSVEP responses and also to evaluate the viewing comfort in each orientation. The signals were processed with band-pass filtering and analysed with Fast Fourier Transform (FFT) and autoregressive spectral analysis. The sign rank statistical results showed horizontal visual stimulus gave the higher response and viewing comfort in all subjects in comparison to vertical orientation.

Products based on binaural beat technology are popular as these claim to relax the user by altering (i.e.entraining) the brain?s neuronal rhythm and a Google search will result in more than a million hits [1]. Whilst this is possible in certain approaches, for example in photic frequency following effects such as in braincomputer interfaces [2]; binaural brain entrainments uses two beats (usually sinusoidal tones) that are different in frequency by a small amount that generates a third pseudo-rhythm at the difference of the two frequencies [3]. Furthermore, binaural brain entrainment is also claimed to allow altered states of consciousness and hemispheric synchronisation [4]. However, the effects of this entrainment on the cardiac rhythm appear to be understudied, though equally important as the effects on the brain. Hence, the investigation here is set out with this aim.

The imagination of limb movements offers an intuitive paradigm for the control of electronic devices via brain computer interfacing (BCI). The analysis of electroencephalographic (EEG) data related to motor imagery potentials has proved to be a difficult task. EEG readings are noisy, and the elicited patterns occur in different parts of the scalp, at different instants and at different frequencies. Wavelet transform has been widely used in the BCI field as it offers temporal and spectral capabilities, although it lacks spatial information. In this study we propose a tailored second generation wavelet to extract features from these three domains. This transform is applied over a graph representation of motor imaginary trials, which encodes temporal and spatial information. This graph is enhanced using per-subject knowledge in order to optimise the spatial relationships among the electrodes, and to improve the filter design. This method improves the performance of classifying different imaginary limb movements maintaining the low computational resources required by the lifting transform over graphs. By using an online dataset we were able to positively assess the feasibility of using the novel method in an online BCI context.

In addition to classic motor signs and symptoms, individuals with Parkinson's disease (PD) are characterized by emotional deficits. Ongoing brain activity can be recorded by electroencephalograph (EEG) to discover the links between emotional states and brain activity. This study utilized machine-learning algorithms to categorize emotional states in PD patients compared with healthy controls (HC) using EEG. Twenty non-demented PD patients and 20 healthy age-, gender-, and education level-matched controls viewed happiness, sadness, fear, anger, surprise, and disgust emotional stimuli while fourteen-channel EEG was being recorded. Multimodal stimulus (combination of audio and visual) was used to evoke the emotions. To classify the EEG-based emotional states and visualize the changes of emotional states over time, this paper compares four kinds of EEG features for emotional state classification and proposes an approach to track the trajectory of emotion changes with manifold learning. From the experimental results using our EEG data set, we found that (a) bispectrum feature is superior to other three kinds of features, namely power spectrum, wavelet packet and nonlinear dynamical analysis; (b) higher frequency bands (alpha, beta and gamma) play a more important role in emotion activities than lower frequency bands (delta and theta) in both groups and; (c) the trajectory of emotion changes can be visualized by reducing subject-independent features with manifold learning. This provides a promising way of implementing visualization of patient's emotional state in real time and leads to a practical system for noninvasive assessment of the emotional impairments associated with neurological disorders.

Non-motor symptoms in Parkinson's disease (PD) involving cognition and emotion have been progressively receiving more attention in recent times. Electroencephalogram (EEG) signals, being an activity of central nervous system, can reflect the underlying true emotional state of a person. This paper presents a computational framework for classifying PD patients compared to healthy controls (HC) using emotional information from the brain's electrical activity.

Parkinson's disease (PD) is not only characterized by its prominent motor symptoms but also associated with disturbances in cognitive and emotional functioning. The objective of the present study was to investigate the influence of emotion processing on inter-hemispheric electroencephalography (EEG) coherence in PD. Multimodal emotional stimuli (happiness, sadness, fear, anger, surprise, and disgust) were presented to 20 PD patients and 30 age-, education level-, and gender-matched healthy controls (HC) while EEG was recorded. Inter-hemispheric coherence was computed from seven homologous EEG electrode pairs (AF3–AF4, F7–F8, F3–F4, FC5–FC6, T7–T8, P7–P8, and O1–O2) for delta, theta, alpha, beta, and gamma frequency bands. In addition, subjective ratings were obtained for a representative of emotional stimuli. Interhemispherically, PD patients showed significantly lower coherence in theta, alpha, beta, and gamma frequency bands than HC during emotion processing. No significant changes were found in the delta frequency band coherence. We also found that PD patients were more impaired in recognizing negative emotions (sadness, fear, anger, and disgust) than relatively positive emotions (happiness and surprise). Behaviorally, PD patients did not show impairment in emotion recognition as measured by subjective ratings. These findings suggest that PD patients may have an impairment of inter-hemispheric functional connectivity (i.e., a decline in cortical connectivity) during emotion processing. This study may increase the awareness of EEG emotional response studies in clinical practice to uncover potential neurophysiologic abnormalities.

Objective: While Parkinson's disease (PD) has traditionally been described as a movement disorder, there is growing evidence of disruption in emotion information processing associated with the disease. The aim of this study was to investigate whether there are specific electroencephalographic (EEG) characteristics that discriminate PD patients and normal controls during emotion information processing. Method: EEG recordings from 14 scalp sites were collected from 20 PD patients and 30 age-matched normal controls. Multimodal (audio-visual) stimuli were presented to evoke specific targeted emotional states such as happiness, sadness, fear, anger, surprise and disgust. Absolute and relative power, frequency and asymmetry measures derived from spectrally analyzed EEGs were subjected to repeated ANOVA measures for group comparisons as well as to discriminate function analysis to examine their utility as classification indices. In addition, subjective ratings were obtained for the used emotional stimuli. Results: Behaviorally, PD patients showed no impairments in emotion recognition as measured by subjective ratings. Compared with normal controls, PD patients evidenced smaller overall relative delta, theta, alpha and beta power, and at bilateral anterior regions smaller absolute theta, alpha, and beta power and higher mean total spectrum frequency across different emotional states. Inter-hemispheric theta, alpha, and beta power asymmetry index differences were noted, with controls exhibiting greater right than left hemisphere activation. Whereas intra-hemispheric alpha power asymmetry reduction was exhibited in patients bilaterally at all regions. Discriminant analysis correctly classified 95.0% of the patients and controls during emotional stimuli. Conclusion: These distributed spectral powers in different frequency bands might provide meaningful information about emotional processing in PD patients.

Brain-computer interfaces (BCI) are useful devices that allow direct control of external devices using thoughts, i.e. brain's electrical activity. There are several BCI paradigms, of which steady state visual evoked potential (SSVEP) is the most commonly used due to its quick response and accuracy. SSVEP stimuli are typically generated by varying the luminance of a target for a set number of frames or display events. Conventionally, SSVEP based BCI paradigms use magnitude (amplitude) information from frequency domain but recently, SSVEP based BCI paradigms have begun to utilize phase information to discriminate between similar frequency targets. This paper will demonstrate that using a single frame to modulate a stimulus may lead to a bi-modal distribution of SSVEP as a consequence of a user attending both transition edges. This incoherence, while of less importance in traditional magnitude domain SSVEP BCIs becomes critical when phase is taken into account. An alternative modulation technique incorporating a 50% duty cycle is also a popular method for generating SSVEP stimuli but has a unimodal distribution due to user's forced attention to a single transition edge. This paper demonstrates that utilizing the second method results in significantly enhanced performance in information transfer rate in a phase discrimination SSVEP based BCI.

Deficits in the ability to process emotions characterize several neuropsychiatric disorders and are traits of Parkinson's disease (PD), and there is need for a method of quantifying emotion, which is currently performed by clinical diagnosis. Electroencephalogram (EEG) signals, being an activity of central nervous system (CNS), can reflect the underlying true emotional state of a person. This study applied machine-learning algorithms to categorize EEG emotional states in PD patients that would classify six basic emotions (happiness and sadness, fear, anger, surprise and disgust) in comparison with healthy controls (HC). Emotional EEG data were recorded from 20 PD patients and 20 healthy age-, education level- and sex-matched controls using multimodal (audio-visual) stimuli. The use of nonlinear features motivated by the higher-order spectra (HOS) has been reported to be a promising approach to classify the emotional states. In this work, we made the comparative study of the performance of k-nearest neighbor (kNN) and support vector machine (SVM) classifiers using the features derived from HOS and from the power spectrum. Analysis of variance (ANOVA) showed that power spectrum and HOS based features were statistically significant among the six emotional states (p < 0.0001). Classification results shows that using the selected HOS based features instead of power spectrum based features provided comparatively better accuracy for all the six classes with an overall accuracy of 70.10% ± 2.83% and 77.29% ± 1.73% for PD patients and HC in beta (13-30 Hz) band using SVM classifier. Besides, PD patients achieved less accuracy in the processing of negative emotions (sadness, fear, anger and disgust) than in processing of positive emotions (happiness, surprise) compared with HC. These results demonstrate the effectiveness of applying machine learning techniques to the classification of emotional states in PD patients in a user independent manner using EEG signals. The accuracy of the system can be improved by investigating the other HOS based features. This study might lead to a practical system for noninvasive assessment of the emotional impairments associated with neurological disorders.

In a previous study, it has been shown that brain activity, i.e. electroencephalogram (EEG) signals, can be used to generate personal identification number (PIN). The method was based on brain–computer interface (BCI) technology using a P300-based BCI approach and showed that a single-channel EEG was sufficient to generate PIN without any error for three subjects. The advantage of this method is obviously its better fraud resistance compared to conventional methods of PIN generation such as entering the numbers using a keypad. Here, we investigate the stability of these EEG signals when used with a neural network classifier, i.e. to investigate the changes in the performance of the method over time. Our results, based on recording conducted over a period of three months, indicate that a single channel is no longer sufficient and a multiple electrode configuration is necessary to maintain acceptable performances. Alternatively, a recording session to retrain the neural network classifier can be conducted on shorter intervals, though practically this might not be viable.

Assessment of awareness for those with disorders of consciousness is a challenging undertaking, due to the complex presentation of the population. Debate surrounds whether behavioral assessments provide greatest accuracy in diagnosis compared to neuro-imaging methods, and despite developments in both, misdiagnosis rates remain high. Music therapy may be effective in the assessment and rehabilitation with this population due to effects of musical stimuli on arousal, attention, and emotion, irrespective of verbal or motor deficits. However, an evidence base is lacking as to which procedures are most effective. To address this, a neurophysiological and behavioral study was undertaken comparing electroencephalogram (EEG), heart rate variability, respiration, and behavioral responses of 20 healthy subjects with 21 individuals in vegetative or minimally conscious states (VS or MCS). Subjects were presented with live preferred music and improvised music entrained to respiration (procedures typically used in music therapy), recordings of disliked music, white noise, and silence. ANOVA tests indicated a range of significant responses (p ? 0.05) across healthy subjects corresponding to arousal and attention in response to preferred music including concurrent increases in respiration rate with globally enhanced EEG power spectra responses (p = 0.05–0.0001) across frequency bandwidths. Whilst physiological responses were heterogeneous across patient cohorts, significant post hoc EEG amplitude increases for stimuli associated with preferred music were found for frontal midline theta in six VS and four MCS subjects, and frontal alpha in three VS and four MCS subjects (p = 0.05–0.0001). Furthermore, behavioral data showed a significantly increased blink rate for preferred music (p = 0.029) within the VS cohort. Two VS cases are presented with concurrent changes (p ? 0.05) across measures indicative of discriminatory responses to both music therapy procedures. A third MCS case study is presented highlighting how more sensitive selective attention may distinguish MCS from VS. The findings suggest that further investigation is warranted to explore the use of music therapy for prognostic indicators, and its potential to support neuroplasticity in rehabilitation programs.

Brain computer interfaces (BCI) provide a new approach to human computer communication, where the control is realised via performing mental tasks such as motor imagery (MI). In this study, we investigate a novel method to automatically segment electroencephalographic (EEG) data within a trial and extract features accordingly in order to improve the performance of MI data classification techniques. A new local discriminant bases (LDB) algorithm using common spatial patterns (CSP) projection as transform function is proposed for automatic trial segmentation. CSP is also used for feature extraction following trial segmentation. This new technique also allows to obtain a more accurate picture of the most relevant temporal–spatial points in the EEG during the MI. The results are compared with other standard temporal segmentation techniques such as sliding window and LDB based on the local cosine transform (LCT).

OBJECTIVE: Multiresolution analysis (MRA) offers a useful framework for signal analysis in the temporal and spectral domains, although commonly employed MRA methods may not be the best approach for brain computer interface (BCI) applications. This study aims to develop a new MRA system for extracting tempo-spatial-spectral features for BCI applications based on wavelet lifting over graphs. APPROACH: This paper proposes a new graph-based transform for wavelet lifting and a tailored simple graph representation for electroencephalography (EEG) data, which results in an MRA system where temporal, spectral and spatial characteristics are used to extract motor imagery features from EEG data. The transformed data is processed within a simple experimental framework to test the classification performance of the new method. MAIN RESULTS: The proposed method can significantly improve the classification results obtained by various wavelet families using the same methodology. Preliminary results using common spatial patterns as feature extraction method show that we can achieve comparable classification accuracy to more sophisticated methodologies. From the analysis of the results we can obtain insights into the pattern development in the EEG data, which provide useful information for feature basis selection and thus for improving classification performance. SIGNIFICANCE: Applying wavelet lifting over graphs is a new approach for handling BCI data. The inherent flexibility of the lifting scheme could lead to new approaches based on the hereby proposed method for further classification performance improvement.

This study addresses two important problem statements, namely, selection of training datasets for online Brain-Computer Interface (BCI) classifier training and determination of participant concentration levels during an experiment. The work also attempted a pilot study to integrate electroencephalograms (EEGs) and Near Infra Red Spectroscopy (NIRS) for possible applications such as the BCI and for measuring cognitive levels. Two experiments are presented, the first being a mathematical task interleaved with rest states using NIRS only. In the next, integration of the EEG-NIRS with reference to P300-based BCI systems as well as the experimental conditions designed to elicit the concentration levels (denoted as ON and OFF states here) during the paradigm, are presented. The first experiment indicates that NIRS can be used to differentiate a concentrated (i.e., mental activity) level from the rest. However, the second experiment reveals statistically significant results using the EEG only. We present details about the equipment used, the participants as well as the signal processing and machine learning techniques implemented to analyse the EEG and NIRS data. After discussing the results, we conclude by describing the research scope as well as the possible pitfalls in this work from a NIRS viewpoint, which presents an opportunity for future research exploration for BCI and cognitive performance measures.

This study investigates the indications of non-linear dynamic structures in electroencephalogram signals. The iterative amplitude adjusted surrogate data method along with seven non-linear test statistics namely the third order autocorrelation, asymmetry due to time reversal, delay vector variance method, correlation dimension, largest Lyapunov exponent, non-linear prediction error and approximate entropy has been used for analysing the EEG data obtained during self paced voluntary finger-movement. The results have demonstrated that there are clear indications of non-linearity in the EEG signals. However the rejection of the null hypothesis of non-linearity rate varied based on different parameter settings demonstrating significance of embedding dimension and time lag parameters for capturing underlying non-linear dynamics in the signals. Across non-linear test statistics, the highest degree of non-linearity was indicated by approximate entropy (APEN) feature regardless of the parameter settings.

Evoked potentials obtained from brain's electrical activity (i.e. electroencephalogram [EEG]) in response to both visual and auditory oddball paradigms have been used in brain-computer interface (BCI) designs, as a means of achieving external control directly from the brain. This is obviously useful for the severely disabled, but it can also be used for creative applications in music performance. We present here a novel approach that combines conventionally used low frequency information with higher frequency in gamma band to enhance the performance of such BCI designs. EEG data were obtained from three and two subjects for the visual and auditory paradigms, respectively. In the visual paradigm, the subjects perceived common pictures like TV, radio, lamp etc., while for the auditory paradigm, the subjects listened to simple computer-generated sounds like "ding", "exclamation", "chimes" etc. Recognition of target picture or sound focused by the subject using the EEG data allows a control mechanism to be designed, which can then be used to control the movement of a wheelchair, to trigger a sound sample, or to create harmonic variations over a synthesised piece of music. The results confirm that using the higher frequency gamma band, which has been mostly overlooked in BCI studies, along with low frequency in the P300 temporal region, give better classification accuracies for both paradigms. This study offers motivation warranting further exploration on the link between low and high frequency of evoked potentials from EEG for use in visual and auditory based BCI designs.

Genetic Algorithms (GAs) were used in a previous study to automate parameter selection for an EEG-based P300-driven Brain-Computer Interface (BCI). The GA approach showed marked improvement over data-insensitive parameter selection; however, it required lengthy execution times thereby rendering it infeasible for online implementation. Automated parameter selection is retained in this work; however, it is achieved using the less computationally intensive N-fold cross-validation (NFCV). Additionally, this study sought to improve BCI classification accuracy using a training data collection and application protocol that the authors refer to as 'Intra-session classifier training and implementation'. Intra-session classifier training and implementation using NFCV-driven automated parameter selection yielded a classification accuracy of 82.94% compared to 45.44% for the inter-session approach using datainsensitive parameters. These findings are significant impact since the intrasession protocol can be applied to any P300-based BCI regardless of its application platform to obtain improved classification accuracy. Copyright © 2012 Inderscience Enterprises Ltd.

Automatic identification of a person's individuality is an important issue today. Brain Computer Interfaces (BCI) which uses EEG as a modality is a promising area for cognitive biometrics. A BCI system could be used to recognise a sequence (say letters, colours or images) by the user. This sequence could form a 'BrainWord', which could be used for authentication in a multimodal environment with other technologies for high security applications. In this work, we studied several variations of the well-known P300 BCI paradigm. The influence of irrelevant stimuli during a task was studied by considering the popular Rapid Serial Visual Paradigm (RSVP) . The variation in spatial locations of the presentation stimuli during a task was studied, by designing a Spatially Varying Paradigm . Comparison of classification accuracies and bit rates for eight participants from a BCI perspective, highlights that RSVP paradigm could be exploited effectively for biometrics.

This paper reports on the development of a proof-of-concept brain-computer music interfacing system (BCMI), which we built to be tested with a patient with Locked-in Syndrome at the Royal Hospital for Neuro-disability, in London. The system uses the Steady State Visual Evoked Potential (SSVEP) method, whereby targets are presented to a user on a computer monitor representing actions available to perform with the system. Each target is encoded by a flashing visual pattern reversing at a unique frequency. In order to make a selection, the user must direct her gaze at the target corresponding to the action she would like to perform. The patient grasped the concept quickly and rapidly demonstrated her skill at controlling the system with minimal practice. She was able to vary the intensity of her gaze, thus changing the amplitude of her EEG and vary the consequent musical parameters. We have proved the concept that such a BCMI system is cost-effective to build, viable, and useful. However, ergonomic and design aspects of the system require further refinement in order to make it more practical for clinical usage. For instance, the system at present requires a therapist to place individual electrodes and calibrate a userâ??s response to each stimulus, which can be time consuming. A new version of the system will require just positioning of a headset and, due to advanced algorithms, will require no calibration. © 2011, The Author(s). All rights reserved.

The steady state visual evoked protocol has recently become a popular paradigm in brain–computer interface (BCI) applications. Typically (regardless of function) these applications offer the user a binary selection of targets that perform correspondingly discrete actions. Such discrete control systems are appropriate for applications that are inherently isolated in nature, such as selecting numbers from a keypad to be dialled or letters from an alphabet to be spelled. However motivation exists for users to employ proportional control methods in intrinsically analogue tasks such as the movement of a mouse pointer. This paper introduces an online BCI in which control of a mouse pointer is directly proportional to a user's intent. Performance is measured over a series of pointer movement tasks and compared to the traditional discrete output approach. Analogue control allowed subjects to move the pointer faster to the cued target location compared to discrete output but suffers more undesired movements overall. Best performance is achieved when combining the threshold to movement of traditional discrete techniques with the range of movement offered by proportional control.

This paper investigates the characterization ability of linear and nonlinear features and proposes combining such features in order to improve the classification of biological signals, in particular single-trial electroencephalogram (EEG) and electrocardiogram (ECG) data. For this purpose, three data sets composed of ECG, epileptic EEG and finger-movement EEG were utilized. The characterization ability of seven nonlinear features namely the approximate entropy, largest Lyapunov exponents, correlation dimension, nonlinear prediction error, Hurst exponent, higher order autocovariance and asymmetry due to time reversal are compared with two linear features namely the autoregressive (AR) reflection coefficients and AR model coefficients. The features were tested by their ability to differentiate between different classes of data using a linear discriminant analysis (LDA) method with tenfold cross-validation. The class separability of combined linear and nonlinear features was assessed using sequential floating forward search with linear discriminant analysis method (SFFS-LDA). The results demonstrated that linear and nonlinear features on their own provided comparable results for the ECG data set and the finger-movement EEG data set whilst the linear features provided a better class separability compared to nonlinear features for the epileptic EEG data set. Combining linear and nonlinear features demonstrated a significant improvement in the class separability for all of the data sets where an average improvement of 20.56% was obtained with the ECG data set, 7.45% with finger-movement data set and 6.62% with the epileptic EEG data set. Overall results suggest that the use of combined linear and nonlinear feature sets would be a better approach for the characterization and classification of biological signals such as EEG and ECG. © 2010 Institute of Physics and Engineering in Medicine.

The presentation order of training patterns to a simplified fuzzy ARTMAP (SFAM) neural network affects the classification performance. The common method to solve this problem is to use several simulations with training patterns presented in random order, where voting strategy is used to compute the final performance. Recently, an ordering method based on min-max clustering was introduced to select the presentation order of training patterns based on a single simulation. In this paper, another single simulation method based on genetic algorithm is proposed to obtain the presentation order of training patterns for improving the performance of SFAM. The proposed method is applied to a 40-class individual classification problem using visual evoked potential signals and three other datasets from UCI repository. The proposed method has the advantages of improved classification performance, smaller network size and lower training time compared to the random ordering and min-max methods. When compared to the random ordering method, the new ordering scheme has the additional advantage of requiring only a single simulation. As the proposed method is general, it can also be applied to a fuzzy ARTMAP neural network when it is used as a classifier. © 2008 Elsevier B.V. All rights reserved.

We present an application level framework which makes use of Wavelet Packet Analysis (WPA) for improved target detection in oddball paradigm, which are being researched for a brain biometric system. The novelty lies in the usage of both P300 (delta and theta band) and gamma band features from a wavelet perspective using just forty trials. The features were extracted using WPA analysis for target detection, wherein Daubechies (Db4) and Coiflet (Coif3) wavelets are used respectively to extract the P300 and Gamma band energy features. A comparison on the classification accuracy is also presented when the P300 features are used with and without Gamma band features. This work also discusses a new dynamic backward referencing technique which seems to enhance the features (delta, theta and gamma band) from eight channels. A Radial Basis Function (RBF) classifier is used to classes the obtained features as target and non-target for both the paradigms. Initial results on these lines from four subjects show motivating results for further time frequency research.

We present various techniques to detect a target in an oddball paradigm. These techniques are presented with reference to two P300 paradigms being studied to build a biometric system using electroencephalogram (EEG) signals. The novel inblock paradigm presented in this paper proposes a possible variant to the known oddball paradigm and analyses the effect of spatial location of the target block with respect to the non-target block. It brings out a new approach in the oddball paradigm, wherein location of the target might help evoke a higher P300 potential. A comparison of the various analysis techniques studied for both the paradigms is also presented. Initial results from four subjects show that energy analysis gave improved results than the traditional amplitude analysis techniques for target detection in the studied oddball paradigms. The results were comparable for all subjects except for one subject where energy analysis provided better target detection, suggesting that energy based methods could be further explored. This proposed novel paradigm is a step towards the online brain biometric system which is being built for authentication in high security scenarios. © 2008 Inderscience Enterprises Ltd.

We present a novel method which uses the P300amplitudes and Gamma band energy for enhanced targetdetection in P300 paradigms which are being researchedtowards a brain biometric system. The novelty of thiswork lies in the usage of both P300 amplitudes alongwith gamma band features which were calculated fromaveraged waveforms using thirty trials (Target and Non-target) for eight channels. P300 amplitudes wascalculated in the 300-600ms range and Gamma bandenergy was computed in the 30-50 Hz using Shannonenergy for target and non-target. A simple decisionmaker was used to classify the obtained feature vectors.Initial results confirm the possibility of using gamma band energy along-with P300 analysis in an oddball paradigm for better classification accuracies. This studyoffers motivation warranting further need to explore thelinks between P300 and Gamma band in P300 paradigms.

Here we present a novel approach to detect P300 wave in single trial Visual Event Related Potential (VERP) signals using improved principal component analysis to enable a faster brain-computer interface (BCI) design. In the process, the principal components (PCs) are selected using novel methods, namely spectral power ratio (SPR) and sandwich spectral power ratio (SSPR). We set out to assess the improved performances of our proposed methods, SPR and SSPR over standard PC selection methods like Kaiser and residual power for speller BCI design. Concluding, the P300 parameters extracted through our proposed SPR and SSPR methods showed improved detection of target characters in the speller BCI. © 2008 Science Publications.

Brain waves are proposed as a biometric for verification of the identities of individuals in a small group. The approach is based on a novel two-stage biometric authentication method that minimizes both false accept error (FAE) and false reject error (FRE). These brain waves (or electroencephalogram (EEG) signals) are recorded while the user performs either one or several thought activities. As different individuals have different thought processes, this idea would be appropriate for individual authentication. In this study, autoregressive coefficients, channel spectral powers, inter-hemispheric channel spectral power differences, inter-hemispheric channel linear complexity and non-linear complexity (approximate entropy) values were used as EEG features by the two-stage authentication method with a modified four fold cross validation procedure. The results indicated that perfect accuracy was obtained, i.e. the FRE and FAE were both zero when the proposed method was tested on five subjects using certain thought activities. This initial study has shown that the combination of the two-stage authentication method with EEG features from thought activities has good potential as a biometric as it is highly resistant to fraud. However, this is only a pilot type of study and further extensive research with more subjects would be necessary to establish the suitability of the proposed method for biometric applications. © 2008 World Scientific Publishing Company.

In this study, we have investigated the electrophysiological differences between alcoholic and control subjects using two different approaches namely complexity and energy (power) analyses. The electroencephalogram data used in this study were recorded from 77 alcoholic and 44 control subjects while the subjects were performing delayed matching to sample object recognition task for three types of stimuli. These were a single stimulus and a second matching or nonmatching stimulus that followed the single stimulus after a delay. The experimental paradigm evokes object recognition, visual short-term memory, and decision-making abilities. The results indicated that all regions (i.e. frontal, central, temporal, parietal, and occipital) in the brain exhibit more complexity and less energy for alcoholic subjects as compared to controls. When different visual stimuli pairs were compared among alcoholic and control subjects, the results from energy analysis showed groupwise differences in occipital and parietal regions. These results provide a strong indication on the impairment in brain's electrophysiological activity for alcoholic subjects due to a history of long-term alcohol abuse. © 2008 Imperial College Press.

The energy of brain potentials evoked during processing of visual stimuli is considered as a new biometric. In particular, we propose several advances in the feature extraction and classification stages. This is achieved by performing spatial data/sensor fusion, whereby the component relevance is investigated by selecting maximum informative (EEG) electrodes (channels) selected by Davies–Bouldin index. For convenience and ease of cognitive processing, in the experiments, simple black and white drawings of common objects are used as visual stimuli. In the classification stage, the Elman neural network is employed to classify the generated EEG energy features. Simulations are conducted by using the hold-out classification strategy on an ensemble of 1,600 raw EEG signals, and 35 maximum informative channels achieved the maximum recognition rate of 98.56?±?1.87%. Overall, this study indicates the enormous potential of the EEG biometrics, especially due to its robustness against fraud.

Electrophysiological impairments of alcoholism have been researched extensively. However, there is none or few reported research on screening methods for chronic alcoholic subjects. Since chronic alcoholics have serious brain dysfunction, a method to screen for them during specific job applications that require good memory, concentration and/or decision making would be useful. In this paper, a method is proposed to discriminate chronic alcoholic from non-alcoholic subjects while they are sober. Energies of electroencephalogram signals in multiple gamma bands recorded while the subjects performed a picture recognition task are used as features by a neural network to detect the chronic alcoholic subjects. Leave one out cross validation strategy reveals that alcoholics could be discriminated from non-alcoholics with accuracy of 94.55%. This pilot study has shown the potential of the method which could be further developed for use in automatic alcoholic screening procedures.

We propose a novel framework to reduce background electroencephalogram (EEG) artifacts from multitrial visual-evoked potentials (VEPs) signals for use in brain-computer interface (BCI) design. An algorithm based on cyclostationary (CS) analysis is introduced to locate the suitable frequency ranges that contain the stimulus-related VEP components. CS technique does not require VEP recordings to be phase locked and exploits the intertrial similarities of the VEP components in the frequency domain. The obtained cyclic frequency spectrum enables detection of VEP frequency band. Next, bandpass or lowpass filtering is performed to reduce the EEG artifacts using these identified frequency ranges. This is followed by overlapping band EEG artifact reduction using genetic algorithm and independent component analysis (G-ICA) which uses mutual information (MI) criterion to separate EEG artifacts from VEP. The CS and GA methods need to be applied only to the training data; for the test data, the knowledge of the cyclic frequency bands and unmixing matrix would be sufficient for enhanced VEP detection. Hence, the framework could be used for online VEP detection. This framework was tested with various datasets and it showed satisfactory results with very few trials. Since the framework is general, it could be applied to the enhancement of evoked potential signals for any application.

The potential of brain electrical activity generated as a response to a visual stimulus is examined in the context of the identification of individuals. Specifically, a framework for the Visual Evoked Potential (VEP)-based biometrics is established, whereby energy features of the gamma band within VEP signals were of particular interest. A rigorous analysis is conducted which unifies and extends results from our previous studies, in particular, with respect to 1) increased bandwidth, 2) spatial averaging, 3) more robust power spectrum features, and 4) improved classification accuracy. Simulation results on a large group of subject support the analysis. © 2007 IEEE.

A novel method for segmentation of heart sounds (HSs) into single cardiac cycle (S1-Systole-S2-Diastole) using homomorphic filtering and K-means clustering is presented. Feature vectors were formed after segmentation by using Daubechies-2 wavelet detail coefficients at the second decomposition level. These feature vectors were then used as input to the neural networks. Grow and Learn (GAL) and Multilayer perceptron-Backpropagation (MLP-BP) neural networks were used for classification of three different HSs (Normal, Systolic murmur and Diastolic murmur). It was observed that the classification performance of GAL was similar to MLP-BP. However, the training and testing times of GAL were lower as compared to MLP-BP. The proposed framework could be a potential solution for automatic analysis of HSs that may be implemented in real time for classification of HSs.

In previous studies, identification of individuals using 61 channel Visual Evoked Potential (VEP) signals from the brain has been shown to be feasible. These studies used neural network classification of gamma band spectral power of VEP signals from 20 individuals. This paper explores our continuing work in this area to include more subjects in the experiment and to reduce the number of required channels using Fisher Discriminant Ratio function. The experimental study showed that 27 optimal channels were sufficient to yield an average classification rate of 90.97% across 800 test VEP patterns from 40 subjects. Being fewer in number than 61 channels, it is less cumbersome, requires lower computational time, design complexity and cost. This was achieved without loss of performance as 61 channels gave an average classification result of 89.11%. The positive results obtained here showed that the neural activity during perception of visual stimulus was different across individuals. This method could be explored further as a biometric tool to identify individuals as the brain signals are difficult to be forged.

This paper investigates the feasibility of using neural network (NN) and late gamma band (LGB) electroencephalogram (EEG) features extracted from the brain to identify the individuality of subjects. The EEG signals were recorded using 61 active electrodes located on the scalp while the subjects perceived a single picture. LGB EEG signals occur with jittering latency of above 280 ms and are not time-locked to the triggering stimuli. Therefore, LGB EEG could only be computed from single trials of EEG signals and the common method of averaging across trials to remove undesired background EEG (i.e. noise) is not possible. Here, principal component analysis has been used to extract single trials of EEG signals. Zero phase Butterworth filter and Parseval's time-frequency equivalence theorem were used to compute the LGB EEG features. These features were then classified by backpropagation and simplified fuzzy ARTMAP NNs into different categories that represent the individuality of the subjects. The results using a tenfold cross validation scheme gave a maximum classification of 97.33% when tested on 800 unseen LGB EEG features from 40 subjects. This pilot investigation showed that the method of identifying the individuality of subjects using NN classification of LGB EEG features is worth further study.

A common method for designing brain-computer Interface (BCI) is to use electroencephalogram (EEG) signals extracted during mental tasks. In these BCI designs, features from EEG such as power and asymmetry ratios from delta, theta, alpha, and beta bands have been used in classifying different mental tasks. In this paper, the performance of the mental task based BCI design is improved by using spectral power and asymmetry ratios from gamma (24-37 Hz) band in addition to the lower frequency bands. In the experimental study, EEG signals extracted during five mental tasks from four subjects were used. Elman neural network (ENN) trained by the resilient backpropagation algorithm was used to classify the power and asymmetry ratios from EEG into different combinations of two mental tasks. The results indicated that 1) the classification performance and training time of the BCI design were improved through the use of additional gamma band features; 2) classification performances were nearly invariant to the number of ENN hidden units or feature extraction method.

In earlier papers, it was shown that recognizing persons using their brain patterns evoked during visual stimulus is possible. In this paper, several modifications are proposed to improve the recognition accuracy. In the method, gamma band spectral power (GBSP) features were computed from the visual evoked potential (VEP) signals recorded from 61 electrodes while subjects perceived a picture. Two methods were used to improve the classification rate. First, principal component analysis (PCA) was used to reduce the noise and background electroencephalogram (EEG) effects from the VEP signals. Second, the GBSP of each channel was normalized by the total GBSP from all the channels. Three classifiers were used: simplified fuzzy ARTMAP (SFA), linear discriminant (LD) and k-nearest neighbor (kNN). The experimental results using 800 VEP signals from 20 subjects with leave-one-out cross-validation strategy showed that PCA improves the classification performance for all the classifiers with normalization giving improved results in certain cases. The best classification performance of 96.50% obtained using the improved method shows that brain signals have suitable biometric properties that could be further exploited.

This paper introduces new hybrid cross-over methods and new hybrid selection methods for real coded genetic algorithm (RCGA), to solve the optimal control problem of a class of hybrid system, which is motivated by the structure of manufacturing environments that integrate process and optimal control. In this framework, the discrete entities have a state characterized by a temporal component whose evolution is described by event-driven dynamics and a physical component whose evolution is described by continuous time-driven systems. The proposed RCGA with hybrid genetic operators can outperform the conventional RCGA and the existing Forward Algorithms for this class of systems. The hybrid genetic operators improve both the quality of the solution and the actual optimum value of the objective function. A typical numerical example of the optimal control problem with the number of jobs varying from 5 to 25 is included to illustrate the efficacy of the proposed algorithm. Several statistical analyses are done to compare the betterment of the proposed algorithm over the conventional RCGA and Forward Algorithm. Hypothesis t-test and Analysis of Variance (ANOVA) test are also carried out to validate the effectiveness of the proposed algorithm. © 2004 Elsevier B.V. All rights reserved.

In this paper, we have designed a two-state brain-computer interface (BCI) using neural network (NN) classification of autoregressive (AR) features from electroencephalogram (EEG) signals extracted during mental tasks. The main purpose of the study is to use Keirn and Aunon's data to investigate the performance of different mental task combinations and different AR features for BCI design for individual subjects. In the experimental study, EEG signals from five mental tasks were recorded from four subjects. Different combinations of two mental tasks were studied for each subject. Six different feature extraction methods were used to extract the features from the EEG signals: AR coefficients computed with Burg's algorithm, AR coefficients computed with a least-squares (LS) algorithm and adaptive autoregressive (AAR) coefficients computed with a least-mean-square (LMS) algorithm. All the methods used order six applied to 125 data points and these three methods were repeated with the same data but with segmentation into five segments in increments of 25 data points. The multilayer perceptron NN trained by the back-propagation algorithm (MLP-BP) and linear discriminant analysis (LDA) were used to classify the computed features into different categories that represent the mental tasks. We compared the classification performances among the six different feature extraction methods. The results showed that sixth-order AR coefficients with the LS algorithm without segmentation gave the best performance (93.10%) using MLP-BP and (97.00%) using LDA. The results also showed that the segmentation and AAR methods are not suitable for this set of EEG signals. We conclude that, for different subjects, the best mental task combinations are different and proper selection of mental tasks and feature extraction methods are essential for the BCI design.

A method of identifying individuals using visual-evoked-potential (VEP) signals and neural network (NN) is proposed. In the approach, a backpropagation (BP) NN is trained to identify individuals using gamma-band (30-50 Hz) spectral power ratio of VEP signals extracted from 61 electrodes located on the scalp of the brain. The gamma-band spectral-power ratio is computed using a zero-phase Butterworth digital filter and Parseval's time-frequency equivalence theorem. NN classification gives an average of 99.06% across 400 test VEP patterns from 20 individuals using 10-fold cross-validation scheme. This shows promise for the approach to be developed further as a biometric identification system.

In this paper, Principal Component Analysis (PCA) is used to reduce noise from multi-channel Visual Evoked Potential (VEP) signals. PCA is applied to reduce noise from multi-channel VEP signals because VEP signals are more correlated from one channel to another as compared to noise during visual perception. Emulated VEP signals contaminated with noise are used to show the noise reduction ability of PCA. These noise reduced VEP signals are analysed in the gamma spectral band to classify alcoholics and non-alcoholics with a Fuzzy ARTMAP (FA) neural network. A zero phase Butterworth digital filter is used to extract gamma band power in spectral range of 30 to 50 Hz from these noise reduced VEP signals. The results using 800 VEP signals give an average FA classification of 92.50% with the application of PCA and 83.33% without the application of PCA. © 2003, The Institute of Electrical Engineers of Japan. All rights reserved.

This paper proposes a new brain-computer interface (BCI) design using fuzzy ARTMAP (FA) neural network, as well as an application of the design. The objective of this BCI-FA design is to classify the best three of the five available mental tasks for each subject using power spectral density (PSD) values of electroencephalogram (EEG) signals. These PSD values are extracted using the Wiener-Khinchine and autoregressive methods. Ten experiments employing different triplets of mental tasks are studied for each subject. The findings show that the average BCI-FA outputs for four subjects gave less than 6% of error using the best triplets of mental tasks identified from the classification performances of FA. This implies that the BCI-FA can be successfully used with a tri-state switching device. As an application, a proposed tri-state Morse code scheme could be utilized to translate the outputs of this BCI-FA design into English letters. In this scheme, the three BCI-FA outputs correspond to a dot and a dash, which are the two basic Morse code alphabets and a space to denote the end (or beginning) of a dot or a dash. The construction of English letters using this tri-state Morse code scheme is determined only by the sequence of mental tasks and is independent of the time duration of each mental task. This is especially useful for constructing letters that are represented as multiple dots or dashes. This combination of BCI-FA design and the tri-state Morse code scheme could be developed as a communication system for paralyzed patients.

A new individual identification technique using visual evoked potential (VEP) signals is proposed. The technique uses fuzzy ARTMAP (FA) classification of gamma band VEP response from 61 channels extracted while the individual is seeing a single picture. All ten subjects tested are classified correctly with an overall maximum classification rate of 95% across 200 test patterns. The method could be developed into a stand-alone identification system or as an addition to existing identification systems, especially for identifying a group of individuals in a company or factory.

In this paper, we propose a technique that uses genetic algorithm (GA) with Fuzzy ARTMAP (FA) classifier to identify the discriminatory subset of the feature set for classification of alcoholics and non-alcoholics using brain rhythm extracted during visual stimulus. In the experimental study, the feature set consists of seven spectral power ratios extracted from 61 visual evoked potential (VEP) channels. The seven spectral bands of VEP signals in the range of 2-50 Hz are extracted using constant gain and uniform bandwidth infinite impulse response (IIR) band-pass filters. Spectral power in these bands are obtained using Parseval's time-frequency energy equivalence theorem. The spectral power ratio for each band is obtained by dividing the spectral power of the band with the total spectral power of the channel. Classification experiments using FA and multilayer perceptron-backpropagation (MLP-BP) classifiers are carried out to confirm that the identified spectral power ratios and channels using the proposed technique are discriminatory. The classification results show that the difference of VEP signals between alcoholics and non-alcoholics can be observed using two spectral power ratios in gamma band (37-50 Hz) extracted from seven channels. This fact indicates that gamma band spectral power could be used to show evidence on the lasting effects of long-term use of alcohol on visual response though the studied alcoholics have been abstinent for a minimum period of 1 month.

In this letter, neural networks (NNs) classify alcoholics and nonalcoholics using features extracted from visual evoked potential (VEP). A genetic algorithm (GA) is used to select the minimum number of channels that maximize classification performance. GA population fitness is evaluated using fuzzy ARTMAP (FA) NN, instead of the widely used multilayer perceptron (MLP). MLP, despite its effective classification, requires long training time (on the order of 103 times compared to FA). This causes it to be unsuitable to be used with GA, especially for on-line training. It is shown empirically that the optimal channel configuration selected by the proposed method is unbiased, i.e., it is optimal not only for FA but also for MLP classification. Therefore, it is proposed that for future experiments, these optimal channels could be considered for applications that involve classification of alcoholics.

In this paper, we are proposing a method to predict cognitive tasks performed by the human brain using spectral analysis of electrical signals extracted from the scalp of the brain. These electrical signals, which are generated by the synapses and neurons in the brain, are also known as Electroencephalogram (EEG) signals. The EEG signals are analysed using autoregressive spectral analysis, a type of modern parametric spectral analysis method, which comparatively yield better power spectrum over the classical Fourier methods. Power spectral densities of the EEG signals are used to train a Fuzzy ARTMAP network to predict the respective cognitive tasks. In our experimental study, we have analysed 3 subjects performing 2 different cognitive tasks and our average results of 72.22% to 93.05% for each subject show that it is highly possible to predict cognitive tasks based on EEG signals. This can be used as a mode of communication or wheelchair control for paralysed patients and also in EEG biofeedback systems.

Generally, visual evoked potential (VEP) estimation requires power spectrum computation. Here VEP is estimated using a recursive digital filter, which is faster and simpler to design. The method assumes VEP can be represented by a 40 Hz spectrum and the filter extracts a 40 Hz signal in the time domain, circumventing the normal procedure of power spectrum computation the normal procedure of power spectrum computation the normal procedure of power spectrum computation. Experimental results validate the method for discriminating alcoholics from control subjects.

The usual regular moment functions are only invariant to image translation, rotation and uniform scaling. These moment invariants are not invariant when an image is scaled non-uniformly in the x- and y-axes directions. This paper addresses this problem by presenting a new technique to obtain moments that are invariant to non-uniform scaling. However, this technique produces a set of features that arc only invariant to translation and uniform/non-uniform scaling. To obtain invariance to rotation, moments are calculated with respect to the x-y-axis of the image. To perform this, a neural network is used to estimate the angle of rotation from the x-y-axis and the image is unrotated to the x-y-axis. Consequently, we are able to obtain features that are invariant to translation, rotation and uniform/non-uniform scaling. The mathematical background behind the development and invariance of the new moments are presented. The results of experimental studies using English alphabets and Arabic numerals scaled uniformly/non-uniformly, rotated and translated are discussed to further verify the validity of the new moments.

Conventional regular moment functions have been proposed as pattern sensitive features in image classification and recognition applications. But conventional regular moments are only invariant to translation, rotation and equal scaling. It is shown that the conventional regular moment invariants remain no longer invariant when the image is scaled unequally in the x- and y-axis directions. We address this problem by presenting a technique to make the regular moment functions invariant to unequal scaling. However, the technique produces a set of features that are only invariant to translation, unequal/equal scaling and reflection. They are not invariant to rotation. To make them invariant to rotation, moments are calculated with respect to the principal axis of the image. To perform this, the exact angle of rotation must be known. But the method of using the second-order moments to determine this angle will also be inclusive of an undesired tilt angle. Therefore, in order to correctly determine the amount of rotation, the tilt angle which differs for different scaling factors in the x- and y-axis directions for the particular image must be obtained. In order to solve this problem, a neural network using the back-propagation learning algorithm is trained to estimate the tilt angle of the image and from this the amount of rotation for the image can be determined. Next, the new moments are derived and a Fuzzy ARTMAP network is used to classify these images into their respective classes. Sets of experiments involving images rotated and scaled unequally in the x- and y-axis directions are carried out to demonstrate the validity of the proposed technique.

The classification of images using regular or geometric moment functions suffers from two major problems. First, odd orders of central moments give zero value for images with symmetry in the x and/or y directions and symmetry at centroid. Secondly, these moments are very sensitive to noise especially for higher order moments. In this paper, a single solution is proposed to solve both these problems. The solution involves the computation of the moments from a reference point other than the image centroid. The new reference centre is selected such that the invariant properties like translation, scaling and rotation are still maintained. In this paper, it is shown that the new proposed moments can solve the symmetrical problem. Next, we show that the new proposed moments are less sensitive to Gaussian and random noise as compared to two different types of regular moments derived by Hu. Extensive experimental study using a neural network classification scheme with these moments as inputs are conducted to verify the proposed method.

Symmetrical images in the x and/or y directions give zero values for odd regular moments, whch cause classification problems. It is proposed that regular moments be computed from centres other than the image centroid, where the centre satisfies the translation, scale, rotation, and/or reflection invariance of regular moments.

Healthcare field is highly benefited by incorporating BCI for detection and diagnosis of some health related detriment as well as rehabilitation and restoration of certain disabilities. An EEG dataset acquired from 15 high-functioning ASD patients, while they were undergoing a P300 experiment in a virtual reality platform, was analysed in this paper using three algorithms. Performance of Bayes Linear Discriminant Analysis (BLDA) was predominant over Convolutional Neural Network (CNN) and Random Undersampling (RUS) Boosting. BLDA rendered 73% overall accuracy in predicting target and the best accuracy for each subject using CNN or BLDA yielded an overall accuracy of 76%.

Understanding how difficult a learning task is for a person allows teaching material to be appropriately designed to suit the person, especially for programming material. A first step for this would be to predict on the task difficulty level. While this is possible through subjective questionnaire, it could lead to misleading outcome and it would be better to do this by tapping the actual thought process in the brain while the subject is performing the task, which can be done using electroencephalogram. We set out on this objective and show that it is possible to predict easy and difficult levels of mental tasks when subjects are attempting to solve Java programming problems. Using a proposed confidence threshold, we obtained a classification performance of 87.05% thereby showing that it is possible to use brain data to determine the teaching material difficulty level which will be useful in educational environments.

Electrical activities from brain (electroencephalogram, EEG) and heart (electrocardiogram, ECG) have been proposed as biometric modalities but the combined use of these signals appear not to have been studied thoroughly. Also, the feature stability of these signals has been a limiting factor for biometric usage. This paper presents results from a pilot study that reveal the combined use of brain and heart modalities provide improved classification performance and further-more, an improvement in the stability of the features over time through the use of binaural brain entrainment. The classification rate was increased, for the case of the neural network classifier from 92.4% to 95.1% and for the case of LDA, from 98.6% to 99.8%. The average standard deviation with binaural brain entrainment using all the inter-session features (from all the subjects) was 1.09, as compared to 1.26 without entrainment. This result suggests the improved stability of both the EEG and ECG features over time and hence resulting in higher classification performance. Overall, the results indicate that combining ECG and EEG gives improved classification performance and that through the use of binaural brain entrainment, both the ECG and EEG features are more stable over time.

Steady state visual evoked potential (SSVEP) is extensively used in the research of brain-computer interface (BCI) and require a controllable and configurable light source. SSVEP requires appropriate control of visual stimulus parameters, such as flicker frequency, light intensity, multi-frequency light source and multi-spectral compositions. Light emitting diodes (LEDs) are extensively used as a light source as they are energy efficient, low power, multi-chromatic, have higher contrast, and support wider frequency ranges. Here, we present the design of a compact versatile visual stimulus which is capable of producing simultaneous multiple frequency RGB LED flicker suitable for a wide range of SSVEP paradigms. The hardware is based upon the open source Arduino platform and supports on-the-fly reprogramming with easily configurable user interface via USB. The design provides fourteen independent high output channels with customisable output voltages. The flicker frequencies can be easily customised within the frequency range of 5–50 Hz, using a look-up table. The LED flickers are generated with single RGB LEDs which generate the required colour or frequency combinations for combined multi-frequency flicker with variable duty cycle to generate SSVEP. Electroencephalogram (EEG) signals have been successfully recorded from five subjects using the stimulator for different frequencies, colours, duty cycle, intensity and multiple frequency RGB source, thereby demonstrating the high usability, adaptability and flexibility of the stimulator. Finally we discuss the possible improvements to the stimulator which could provide real time user feedback to reduce visual fatigue and so increase the level of user comfort.

Electroencephalogram (EEG) signals are useful for diagnosing various mental conditions such as epilepsy, memory impairments and sleep disorders. Brain-Computer Interface (BCI) is a revolutionary new area using EEG that is most use-ful for the severely disabled individuals for hands-off device control and commu-nication as they create a direct interface from the brain to the external environ-ment, therefore circumventing the use of peripheral muscles and limbs. However, being non-invasive, BCI designs are not necessarily limited to this user group and other applications for gaming, music, biometrics etc have been developed more recently. This chapter will give an introduction to EEG based BCI and existing methodologies; specifically those based on transient and steady state evoked po-tentials, mental tasks and motor imagery will be described. Two real-life scenarios of EEG based BCI applications in biometrics and device control will also be brief-ly explored. Finally, current challenges and future trends of this technology will be summarized.

Electrooculogram (EOG) signals have been used in designing Human-Computer Interfaces, though not as popularly as electroencephalogram (EEG) or electromyogram (EMG) signals. This paper explores several strategies for improving the analysis of EOG signals. This article explores its utilization for the extraction of features from EOG signals compared with parametric, frequency-based approach using an autoregressive (AR) model as well as template matching as a time based method. The results indicate that parametric AR modeling using the Burg method, which does not retain the phase information, gives poor class separation. Conversely, the projection on the approximation space of the fourth level of Haar wavelet decomposition yields feature sets that enhance the class separation. Furthermore, for this method the number of dimensions in the feature space is much reduced as compared to template matching, which makes it much more efficient in terms of computation. This paper also reports on an example application utilizing wavelet decomposition and the Linear Discriminant Analysis (LDA) for classification, which was implemented and evaluated successfully. In this application, a virtual keyboard acts as the front-end for user interactions.

This paper investigates a method to generate personal identification number (PIN) using brain activity recorded from a single active electroencephalogram (EEG) channel. EEG based biometric to generate PIN is less prone to fraud and the method is based on the recent developments in brain-computer interface (BCI) technology, specifically P300 based BCI designs. Our perfect classification accuracies from three subjects indicate promise for generating PIN using thought activity measured from a single channel.

In the last several decades, computers or automated technologies have been utilized to verify the identity of humans using biometrics (i.e., physical and behavioral characteristics) (Wayman, Jain, Maltoni, & Maio, 2004), as it often surpasses the conventional automatic identity verification measures like passwords and personal identification numbers (PINs) by offering positive human identification. For example, the use of a PIN actually denotes the automatic identification of the PIN, not necessarily identification of the person who has provided it. The same applies with cards and tokens, which could be presented by anyone who successfully steals the card or token. PINs and passwords also have the problem of being compromised by 'shoulder surfing' and people picking the obvious choices. Even the recently proposed graphical passwords share similar problems.

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. Therefore, in this paper, EEG signals from imagined activities are proposed as a biometric to identify the individuality of persons. The approach is based on the classification of EEG signals recorded when a user performs either one or several mental activities (up to five). As different individuals have different thought processes, this idea would be appropriate for individual identification. To increase the inter-subject differences, EEG data from six electrodes are used instead of one. A total of 108 features (autoregressive coefficients, channel spectral powers, inter-hemispheric channel spectral power differences and inter-hemispheric channel linear complexity values) are computed from each EEG segment for each mental activity and classified by a linear discriminant classifier using a modified 10 fold cross validation procedure, which gave perfect classification when tested on 500 EEG patterns from five subjects. This initial study has shown the huge potential of the method over existing biometric identification systems as it is impossible to be faked.

We further explore the possibility of using the energy of brain potentials evoked during processing of visual stimuli (VS) as a new biometric tool, where biometric features representing the energy of high frequency electroen-cephalogram (EEG) spectra are used in the person identification paradigm. For convenience and ease of processing of cognitive processing, in the experiments, simple black and white drawings of common objects are used as VS. In the classification stage, the Elman neural network is employed to classify the generated EEG features. The high recognition rate of 99.62% on an ensemble of 800 raw EEG signals indicates the potential of the proposed method.

The advantages of autoregressive (AR) modelling over the classical Fourier Transform methods have been centre staged in the recent years. But a problem with AR method lies with the appropriate model order selection. In this paper, we address this problem by studying the performance of three different types of order selection criteria for AR models to represent electroencephalogram signals. We perform this by extracting EEG signals for different mental tasks and obtaining the appropriate model order given by the different criteria. From this, we derive the spectral density function. Using the spectral values, we train a neural network and classify the tasks into their respective categories. In this way, we show the difference in the performance level of the different model order selection criteria for EEG signals.

Analysis of heart rate variability (HRV) can reveal a range of useful information regarding the dynamics of the autonomic nervous system (ANS). It is considered a robust and reliable tool to understand even some subtle changes in ANS activity. Here, we study the "hidden" characteristic changes in HRV during visually induced motion sickness; using nonlinear analytical methods, supplemented by conventional time- and frequency-domain analyses. We computed HRV from electrocardiograms (ECG) of 14 healthy participants measured at baseline and during nausea. Primarily hypothesizing evident differences in measures of physiologic complexity (SampEn; sample entropy, FuzzyEn; fuzzy entropy), chaos (LLE; largest Lyapunov exponent) and Poincare ́/Lorenz (CSI; cardiac sympathetic activity, CVI; cardiac vagal index) between the two states. We found that during nausea, participants showed a markedly higher degree of regularity (SampEn, p = 0.0275; FuzzyEn, p = 0.0006), with a less chaotic ANS response (LLE, p = 0.0004). CSI significantly increased during nausea compared to baseline (p = 0.0005), whereas CVI did not appear to be statistically different between the two states (p = 0.182). Our findings suggest that motion sickness-induced ANS perturbations may be quantifiable via nonlinear HRV indices. These findings have implications for understanding the malaise of motion sickness and in turn, aid development of therapeutic interventions to relieve motion sickness symptoms.

Transcutaneous auricular vagus nerve stimulation (taVNS) is a novel neuromodulation application for vagal afferent stimulation. Owing to its non-invasive nature, taVNS is a potent therapeutic tool for a diverse array of diseases and disorders that ail us. Herein, we investigated taVNS-induced effects on neural activity of participants during visually induced motion sickness. 64-channel electroencephalography (EEG) recordings were obtained from 15 healthy participants in a randomized, within-subjects, cross-over design during sham and taVNS conditions. To assess motion sickness severity, we used the motion sickness assessment questionnaire (MSAQ). We observed that taVNS attenuated theta (4-8 Hz) brain activity in the right frontal, right parietal and occipital cortices when compared to sham condition. The total MSAQ scores, and central, peripheral and sopite MSAQ categorical scores were significantly lower after taVNS compared to sham. These findings reveal for the first time the potential therapeutic role of taVNS toward counter-motion sickness, and suggest that taVNS may be reliable in alleviating symptoms of motion sickness in real-time, non-pharmacologically.

This study examines the neural activities of participants undergoing vibro-motor reprocessing therapy (VRT) while experiencing motion sickness. We evaluated the efficacy of vibro-motor reprocessing therapy, a novel therapeutic technique based on eye movement desensitization and reprocessing (EMDR), in reducing motion sickness. Based on visually induced motion sickness in two sets of performed sessions, eight participants were exposed to VRT stimulation in a VRT/non-VRT setting. Simultaneously, brain activity changes were recorded using electroencephalography (EEG) at baseline and during stimulus exposure, and comparisons made across the VRT/non-VRT conditions. A significant reduction in the alpha (8-12 Hz) spectral power was observed in the frontal and occipital locations, consistent across all participants. Furthermore, significant reductions were also found in the frontal and occipital delta (0.5-4 Hz) and theta (4-8 Hz) spectral power frequency bands between non-VRT and VRT conditions (p < 0.05). Our results offer novel insights for a potential nonpharmacological treatment and attenuation of motion sickness. Furthermore, symptoms can be observed, and alleviated, in real-time using the reported techniques.

Food variety influences appetitive behaviour, motivation to eat and energy intake. Research found that repeated exposure to varied food images increases the motivation towards food in adults and children. This study investigates the effects of repetition on the modulation of early and late components of event-related potentials (ERPs) when participants passively viewed the same food and non-food images repeatedly. The motivational attention to food and non-food images were assessed in frontal, centroparietal, parietooccipital and occipitotemporal areas of the brain. Participants showed increased late positive potential (late ERP component) to high caloric image in the occipitotemporal region compared to low caloric and nonfood images. Similar effects could be seen in the early ERP component in the frontal region, but with reversed polarity. Data suggest that both the early and late ERP components show greater ERP amplitude when viewing high caloric images than low caloric and non-food images. Despite repeated exposure to same image, high caloric food continued to show sustained attention compared to low caloric and non-food image.

Passive detection of footsteps in domestic settings can allow the development of assistive technologies that can monitor mobility patterns of older adults in their home environment. Acoustic footstep detection is a promising approach for nonintrusive detection of footsteps. So far there has been limited work in developing robust acoustic footstep detection systems that can operate in noisy home environments. In this paper, we propose a novel application of the Attention based Recurrent Deep Neural Network to detect human footsteps in noisy overlapping audio streams. The model is trained on synthetic data which simulates the acoustic scene in a home environment. To evaluate performance, we reproduced two footstep detection models from literature and compared them using the newly developed Polyphonic Sound Detection Scores (PSDS). Our model achieved the highest PSDS and is close to the highest score achieved by generic indoor AED models in DCASE. The proposed system is designed to both detect and track footsteps within a home setting, and to enhance state-of-the-art digital health-care solutions for empowering older adults to live autonomously in their own homes.

In order to reduce motion sickness (MS), a novel vibro-motor reprocessing therapy (VRT), based on an Eye Movement Desensitization and Reprocessing (EMDR) technique is investigated. To the best of the authors knowledge, this is the first time that reprocessing therapy has been evaluated for alleviating MS. Experimentally, MS was induced using visual stimulus of motion videos. Subjective MS was then recorded at baseline for both VRT and non-VRT stimulation conditions, and after each condition, evaluated using a Motion Sickness Assessment Questionnaire (MSAQ). MSAQ scores were compared for both conditions in eight test subjects, with a significant and clear reduction in motion sickness symptoms revealed when applying VRT stimulation. While the subject pool is small, this pilot study indicates that the proposed approach has potential for future exploration in terms of non-pharmacological treatment and management of MS.

Modern audiometry is largely a behavioural task, with the pure-tone audiogram (PTA) being the gold standard for evaluating frequency-specific hearing thresholds in adults. The nature of behavioural audiometry makes estimating accurate hearing thresholds difficult in infants and people with disabilities, where following instructions or interacting with the test may be difficult or impossible. We propose a method in which Auditory Evoked Potentials (AEPs) are used as an alternative to behavioural audiometry for detecting frequency-specific thresholds. Specifically, P300 responses elicited by the tones of a PTA are automatically detected from electroencephalogram (EEG) data, to evaluate hearing acuity. To assess the effectiveness of this method, we created a dataset of EEG recordings from participants presented with a series of pure tones at 6 different frequencies with steadily decreasing volumes, during a PTA test. This dataset was used to train a support vector machine (SVM) to identify when a participant was played a tone, whether they perceived it or not using their EEG. Results demonstrate that detecting hearing events can be very accurate for participants for whom the classifier has been trained apriori. However, accuracy drops significantly for unseen participants - when the classifier has not been trained on any prior data from a given participant before classifying their EEG. However, by establishing that AEP response-based audiometry is viable for detecting tones, future work will explore the ability of more powerful deep neural networks to accurately estimate for unseen participants.

This study explores auricular vagus nerve stimulation (aVNS) within the context of stress. Five healthy subjects underwent a pulsed mode aVNS (with a frequency of 25 Hz and pulse width of 200 μs) using a custom made current stimulation device. The device triggered the auricular vagus nerve branch through the tragus for 15 minutes, with prefrontal EEG data collected pre and post aVNS (each for 5 minutes) to capture stress indicators. Given that the frontal location is prone to noise, EEG signal pre-processing through independent component analysis (ICA) was used to reduce eye movements and other artifacts. Alpha (8-14 Hz) asymmetry in prefrontal location in both hemispheres was computed from locations Fp1 and Fp2 for one second segments of EEG; these asymmetry values were normalised by subtracting the post aVNS (or placebo) from pre aVNS (placebo). The normalised alpha frontal asymmetries were assessed for statistical significance using T-test. Results showed increased aVNS alpha prefrontal asymmetry compared to placebo, indicating reduced stress. This was statistically significant ( ) for all five subjects. It is therefore concluded that the pulsed mode aVNS stimulation at the tragus may represent a potential approach for stress relief.

Non-invasive vagus nerve stimulation is fast becoming a popular alternative treatment method for various health disorders. The authors investigated the effects of auricular vagus nerve stimulation at tragus for activating the parasympathetic nervous system to reduce stress, in light of mixed results from other studies. Stimulation frequency of 25 Hz with a pulse-width of 200 μs was administered at tragus with ECG data recorded during pre- and post-stimulation trials to investigate changes in the low-frequency (LF) and high-frequency (HF) parameters of heart rate variability (HRV). The results from five subjects demonstrate an increase in the HF component and a decrease in LF when comparing pre- and post- stimulation values, denoting that VNS stimulated more of the parasympathetic activity. The LF/HF ratio was reduced for all participants after stimulation, with an average reduction of 64:5% observed. Overall, this study has indicated the feasibility of using tragus as a stimulation site to stimulate the vagus nerve; tragus being easier to administrate than many alternative sites while still being effective for stress reduction.

In this work, we propose an approach that features deep feature embedding learning and hierarchical classification with triplet loss function for Acoustic Scene Classification (ASC). In the one hand, a deep convolutional neural network is firstly trained to learn a feature embedding from scene audio signals. Via the trained convolutional neural network, the learned embedding embeds an input into the embedding feature space and transforms it into a high-level feature vector for representation. In the other hand, in order to exploit the structure of the scene categories, the original scene classification problem is structured into a hierarchy where similar categories are grouped into meta-categories. Then, hierarchical classification is accomplished using deep neural network classifiers associated with triplet loss function. Our experiments show that the proposed system achieves good performance on both the DCASE 2018 Task 1A and 1B datasets, resulting in accuracy gains of 15.6% and 16.6% absolute over the DCASE 2018 baseline on Task 1A and 1B, respectively.

Recently, the subject-specific surface electromyography (sEMG)-based gesture classification with deep learning algorithms has been widely researched. However, it is not practical to obtain the training data by requiring a user to perform hand gestures many times in real life. This problem can be alleviated to a certain extent if sEMG from many other subjects could be used to train the classifier. In this paper, we propose a normalisation approach that allows implementing real-time subject-independent sEMG based hand gesture classification without training the deep learning algorithm subject specifically. We hypothesed that the amplitude ranges of sEMG across channels between forearm muscle contractions for a hand gesture recorded in the same condition do not vary significantly within each individual. Therefore, the min-max normalisation is applied to source domain data but the new maximum and minimum values of each channel used to restrict the amplitude range are calculated from a trial cycle of a new user (target domain) and assigned by the class label. A convolutional neural network (ConvNet) trained with the normalised data achieved an average 87.03 accuracy on our G. dataset (12 gestures) and 94.53 on M. dataset (7 gestures) by using the leave-one-subject-out cross-validation.

This paper presents a robust deep learning framework developed to detect respiratory diseases from recordings of respiratory sounds. The complete detection process firstly involves front end feature extraction where recordings are transformed into spectrograms that convey both spectral and temporal information. Then a back-end deep learning model classifies the features into classes of respiratory disease or anomaly. Experiments, conducted over the ICBHI benchmark dataset of respiratory sounds, evaluate the ability of the framework to classify sounds. Two main contributions are made in this paper. Firstly, we provide an extensive analysis of how factors such as respiratory cycle length, time resolution, and network architecture, affect final prediction accuracy. Secondly, a novel deep learning based framework is proposed for detection of respiratory diseases and shown to perform extremely well compared to state of the art methods.

Acoustic scene classification (ASC) using front-end timefrequency features and back-end neural network classifiers has demonstrated good performance in recent years. However a profusion of systems has arisen to suit different tasks and datasets, utilising different feature and classifier types. This paper aims at a robust framework that can explore and utilise a range of different time-frequency features and neural networks, either singly or merged, to achieve good classification performance. In particular, we exploit three different types of frontend time-frequency feature; log energy Mel filter, Gammatone filter and constant Q transform. At the back-end we evaluate effective a two-stage model that exploits a Convolutional Neural Network for pre-trained feature extraction, followed by Deep Neural Network classifiers as a post-trained feature adaptation model and classifier. We also explore the use of a data augmentation technique for these features that effectively generates a variety of intermediate data, reinforcing model learning abilities, particularly for marginal cases. We assess performance on the DCASE2016 dataset, demonstrating good classification accuracies exceeding 90%, significantly outperforming the DCASE2016 baseline and highly competitive compared to state-of-the-art systems.

This work proposes bag-of-features deep learning models for acoustic scene classi?cation (ASC) – identifying recording locations by analyzing background sound. We explore the effect on classi?cation accuracy of various front-end feature extraction techniques, ensembles of audio channels, and patch sizes from three kinds of spectrogram. The back-end process presents a two-stage learning model with a pre-trained CNN (preCNN) and a post-trained DNN (postDNN). Additionally, data augmentation using the mixup technique is investigated for both the pre-trained and post-trained processes, to improve classi?cation accuracy through increasing class boundary training conditions. Our experiments on the 2018 Challenge on Detection and Classi?cation of Acoustic Scenes and Events - Acoustic Scene Classi?cation (DCASE2018-ASC) subtask 1A and 1B signi?cantly outperform the DCASE2018 reference implementation and approach state-of-the-art performance for each task. Results reveal that the ensemble of multi-spectrogram features and data augmentation is bene?cial to performance.

Physiological measurement like surface electromyography (sEMG) allows a deeper insight on interactions among subsystems during the human motion coordination. In this paper, we aim to investigate such interactions via functional muscle networks during hand movements, especially when different hand gestures are performed. It is achieved by considering muscle connectivities using Granger Prediction of paired sEMG signals, which were recorded from extrinsic muscles of the upper limb, while participants were sitting upright and performing hand gestures. It is found that by using muscle connectivities obtained by applying the method of Granger Prediction as features, although individual differences exist among subjects, significant connections between pairs of muscles were observed through permutation tests at a group level. Graph theory based on the overall statistical result was used to visualise functional networks by considering all the significant connections which were not bidirectional. We found two distinct networks can be used to represent the differences between two hand gestures. Such insight of functional networks can be a potential candidate to interpret the relationships between muscle pairs, which is helpful for decoding hand gestures.

Sound event detection is an extension of the static auditory classification task into continuous environments, where performance depends jointly upon the detection of overlapping events and their correct classification. Several approaches have been published to date which either develop novel classifiers or employ well-trained static classifiers with a detection front-end. This paper takes the latter approach, by combining a proven CNN classifier acting on spectrogram image features, with time-frequency shaped energy detection that identifies seed regions within the spectrogram that are characteristic of auditory energy events. Furthermore, the shape detector is optimised to allow early detection of events as they are developing. Since some sound events naturally have longer durations than others, waiting until completion of entire events before classification may not be practical in a deployed system. The early detection capability of the system is thus evaluated for the classification of partial events. Performance for continuous event detection is shown to be good, with accuracy being maintained well when detecting partial events.

Existing research on task difficulty and program comprehension mainly concentrate on brain areas related to attention and meditation. In this research, an in-depth analysis of Task Difficulty Level (TDL) for program comprehension is proposed with features extracted from different areas of the brain. Two levels of task difficulty were analysed: easy and difficult. Eight students were asked to solve nine Java programs of different difficulty level and the subject's cognitive load was recorded using EEG. Four different feature extraction methods were used for analysis - Energy, Frequency ratio, Event Related De-Synchronization (ERD) and Asymmetry ratio and Naïve Bayes classifier was used for classifying different TDL. The results indicated that the recorded EEG signals could reflect TDL for program comprehension tasks for predicting easy and difficult tasks. The classifier predicted task difficulty (easy/difficult) of a new task with an overall correctly classified accuracy of 76.55% with a precision of 80.03% and recall of 76.66%. This in-depth analysis of TDL for program comprehension tasks using EEG could support in developing future generation learning tools.

This paper investigates the possibilities of developing a hybrid brain-computer interface based on Steady State Visual Evoked Potential (SSVEP) and P300 responses. SSVEP classification accuracy is improved using P300 event detection as a secondary validation technique in this study. SSVEP events are generated using a hybrid visual stimuli consisting of four independent radial chip-on-board green LED rings flashing at frequencies 7, 8 9 and 10 Hz, which are controlled by four 32-bit microcontrollers to ensure precise generation of flashing frequencies. P300 events are generated with a flash stimulus controller that produces random red LED flashes using high power single LED located inside each of the four radial rings. The P300 flashes are marked as events along with the recorded SSVEP EEG. The study analysed the EEG data recorded from five participants comprising of five trials each, which included both SSVEP and P300 events to identify the classification effectiveness for hybrid BCI. The EEG data was band-pass filtered and events extracted using custom MATLAB algorithms showed that SSVEP classifications could be improved using P300 events for reliable BCI applications. © 2017 IEEE.

This study investigates the influence of LED visual stimulus brightness in evoking steady state visual evoked potential (SSVEP) responses in brain which can be utilised for vision research, medical diagnostics or for developing brain-computer interfaces (BCI). LED visual stimulus was based on a radial 130 mm chip on board (COB) LEDs emitting green light. The frequencies of the flickers were precisely controlled by a 32-bit microcontroller platform to generate SSVEP with high accuracy. For this study the luminosity of the visual stimulus was controlled externally at levels of 25, 50, 75 and 100% of the maximum visual stimulus brightness. The SSVEP frequencies used to investigate the luminosity effect were 7, 8, 9 and 10 Hz for a period of 30 seconds for all the four luminous levels of the visual stimulus. The study analysed the EEG recordings from five participants comprising of five trials for each frequency and luminous levels of visual stimulus. The results indicated that luminosity at 75% of the full brightness gave significantly higher response for all participants when compared to other luminous levels of the visual stimulus with reduced visual fatigue. © 2016 School of Mechanical and Systems Engineering.

This study proposes the use of radial visual stimuli design to obtain increased steady state visual evoked potential (SSVEP) responses that can be utilised in brain-computer interfaces (BCI). Visual stimuli designs based on chip on board (COB) LEDs were used in this study to compare the influences of the radial with horizontal and concentric patterns in SSVEP. Circular rings with diameters 70 mm, 90 mm, 110 mm, and 130 mm with green COB LEDs were used for radial and concentric patterns while green strip COB LED of 18 mm width and length 16 cm was used for horizontal pattern. The visual flicker and the concentric patterns were generated and controlled precisely by a 32-bit microcontroller platform. The SSVEP frequencies used were 7, 8, 9 and 10 Hz for a period of 30 seconds for each horizontal, radial and concentric circle visual stimulus. The study analysed the EEG recording from five participants comprising of five trials from each frequency and three different stimuli designs to identify the most responsive visual stimulus for evoking SSVEP. Furthermore, we also compared the influence of ring diameters in radial visual stimulus to identify the maximal response and minimal visual fatigue. The results indicated that radial stimulus gave significantly better response than concentric circles or horizontal stimuli for all the participants. Also 130 mm radial stimulus gave the highest response and viewing comfort.

It is always desirable to have an accurate system that allows fast recovery of patients undergoing physiotherapy in terms of integrated health and cost benefits. The caregivers and medical personnel too gain a lot of expertise through the innovations involved in treatment methodology. This system proposed here was developed successfully with a straightforward Segmented-Mean feature construction method that enables its portability to suit smart biomedical devices. In this work, four different exercises were completed by four different subjects in two sessions and the feedback system was generated from every single trial performance via a visual display in a smart phone. The accuracy of the system's output depends on the precise representation of two important things namely, correct gesture and timings. These two parameters have to be captured from the signals that are generated by the hand glove during the manual physiotherapy as guided by the experts during the teaching (i.e. Training) phase. Any deviation from the model should also be captured and reflected in the feedback to align the physio-movements towards perfection to minimise adverse effects. So the feature has to be constructed with complete representation and obviously, as fast as possible. The proposed Segmented-Mean method calculates the mean of data that arrives from the significant electrodes periodically, thus preserving the performance of the subject and is found suitable in estimating the enactment of exercises and required deviations (if any), accurately and as appropriate. The proposed Segmented-Mean method helps the construction of features easily than other conventional methods by reducing the computational complexity and therefore, the response time. Hence, shifting the importance to actual physiotherapy monitoring with an accurate system that works on simple feature construction made feasible.

In this paper, we explore the visual effects of animated 2D line strokes and 3D cubes. A given 2D image is segmented into either 2D line strokes or 3D cubes. Each segmented object (i.e., line stroke or each cube) is initialised with the position and the colour of the corresponding pixel in the image. The program animates these objects using the boid framework. This simulates a flocking behavior of line strokes in a 2D space and cubes in a 3D space. In this implementation the animation runs in a cycle from the disintegration of the original image to a swarm of line strokes or 3D cubes, then the swarm moves about and then integrates back into the original image (an example clip has been uploaded to YouTube and can be viewed at https://www.youtube.com/watch?v=aV6h0VzTZ8).

Most LED visual stimulators used in steady state visual evoked potential (SSVEP) brain-computer interface (BCI) use single LED sources to elicit SSVEP responses. In this study, we tested the hypothesis that different orientations would have different responses in different participants and aimed to develop a portable LED based stimulus design which consists of a small number of RGB LEDs arranged in a line which can be oriented horizontally or vertically. The colour and frequency of the flicker were controlled by a portable microcontroller platform. The study investigated the performance of the SSVEP from five participants when the LED stimulus was displayed vertically and horizontally for a period of 30 seconds. The frequency range used was from 7 Hz to 11 Hz with three primary colours: red, green and blue in both orientations. Furthermore, we also compared the effect of vertical and horizontal orientations using four different frequencies and three colours to test visual fatigue reduction. The results of the analysis using band-pass filtering and Fast Fourier Transform showed that the green horizontal LED stimulus orientation gave the highest response and viewing comfort in all the participants rather than the vertical orientation.

In this paper, it is shown that classification of features from heart (electrocardiogram, ECG) signals for biometric purposes (i.e. for individual identification) degrades over a period of time and a method based on binaural brain entrainment is proposed to minimise the variations in the heart signals over time to improve the classification performance. The results indicate that variability of the heart features is reduced by 15.57% using the proposed method and this results in improving the classification accuracy from 90.35% to 95.77% when tested with five subjects with ECG data recorded over a period of six months. This pilot study indicates that binaural brain entrainment can be used to improve the stability of ECG features over time thereby increasing its potential to be used in biometric applications. © 2014 IEEE.

Steady state visually evoked potentials (SSVEP) are extensively used in the research of brain-computer interface (BCI) and require a configurable light source flashing at different frequencies. Precise control of simultaneous multiple frequencies are essential for SSVEP studies and also for reducing the visual fatigue. Instead of LCD based stimulus which requires more resources and power, light emitting diodes (LEDs) are used as a light source as they are energy efficient, consume lower power, have higher contrast, less tiring visually, have multi-chromatic function and supports wider frequency ranges. In this paper, we propose a visual stimulator using off-shelf components to build a simple and yet customisable LED stimulus for testing the performance and qualitative user comfort using SSVEP electroencephalogram (EEG) data.

Brain computer interfaces are control systems that allow the interaction with electronic devices by analysing the user's brain activity. The analysis of brain signals, more concretely, electroencephalographic data, represents a big challenge due to its noisy and low amplitude nature. Many researchers in the field have applied wavelet transform in order to leverage the signal analysis benefiting from its temporal and spectral capabilities. In this study we make use of the so-called second generation wavelets to extract features from temporal, spatial and spectral domains. The complete multiresolution analysis operates over an enhanced graph representation of motor imaginary trials, which uses per-subject knowledge to optimise the spatial links among the electrodes and to improve the filter design. As a result we obtain a novel method that improves the performance of classifying different imaginary limb movements without compromising the low computational resources used by lifting transform over graphs.

Among the many paradigms used in brain-computer interface (BCI), steady state visual evoked potential (SSVEP) offers the quickest response; however it is disadvantageous from the point of view of visual fatigue, which prevents subjects from prolonged usage of visual stimuli especially when LEDs are used. In this paper, we propose a visual stimulator using readily available RGB LEDs with clear and frosted glass, with the latter being tested for performance and qualitative user comfort using electroencephalogram (EEG) data from four subjects. Furthermore, we also compare frosted and clear stimuli for three colours Red, Green and Blue with frequency values of 7, 8, 9 and 10 Hz. The results using band-pass filtering and Fast Fourier Transform showed that 7 Hz Green clear LED stimuli gave the highest response in general, although all the subjects indicated that they were more comfortable with frosted LED stimuli.

Brain computer interfacing (BCI) offers the possibility to interact with machines uniquely relying on the user's thoughts. Although wavelet analysis has been used in the BCI field there is evidence that standard wavelet families, such as Daubechies, may not be the optimal approach. In this study, we developed a novel wavelet lifting scheme, specifically for BCI design. The lifting transform in this new approach is based on common spatial patterns (CSP), which allows to exploit the signal characteristics in temporal, spectral and spatial domains simultaneously. Experimental results show that in BCI applications the new wavelet outperforms several first generation wavelet families in terms of classification accuracy and resource consumption.

Single cycle analysis of the Steady State Visual Evoked Potential (SSVEP) response allows use of standard significance tests to determine a subject's gaze. This allows standard criterion levels to be set across all subjects rather than individually by empirical analysis. Furthermore once a subject's gaze has been classified it is possible to extract a notion of expected phase to a SSVEP stimulus. This expected phase value can then be used to augment classification of attention to a four class Brain Computer Interface system improving transfer rates.

The use of both linear autoregressive model coefficients and nonlinear measures for classification of EEG signals recorded from healthy subjects and epilepsy patients is investigated. A total of seven nonlinear measures namely the approximate entropy, largest lyapunov exponent, correlation dimension, nonlinear prediction error, hurst exponent, third order autocovariance, asymmetry due to time reversal, are used in this study. The class separability of individual and combined feature sets is measured using Linear Discriminant Analysis (LDA) algorithm where the multiple features are selected by sequential floating forward search (SFFS) algorithm. The results have shown that the use of combined feature sets provide a better characterization of EEG signals compared to individual features.

The performance of any brain-computer interface (BCI) highly depends on being artifact free. In this study, we propose a mathematical modelling approach to design an efficient non-invasive BCI based on P300 component found in single trial visual evoked potential (VEP) signals. Since the brain processes multiple functions simultaneously the extracted VEP results are in a complex pattern. Further, the characteristics of the P300 component are difficult to be determined a priori especially when the signals are analysed on single trial basis. However, the data used by BCI systems have high dimensionality due to the recording from multiple electrode locations and this high dimensionality could be exploited for reducing the effects from artifacts, using specific pre-processing techniques. In this research study, we propose a mathematical framework for noise reduction and a two-step classification using dynamic methods that results in an enhanced BCI design. The application of singular value decomposition (SVD) to the discrete single trial VEP data facilitates reduction of noise and operational data dimension. Frequency specific filtering further reduces noise and a computationally simple distance based measure with novel method of using two thresholds was utilised for classification. The experimental results give a very low false accept rate (FAR) and false reject rate (FRR) and a near negligible equal error rate (EER) of 2.91%. The high accuracy obtained validates our proposed single trial based approach.

Genetic algorithm (GA) is used for jointly estimating the embedding dimension and time lag parameters in order to achieve an optimal reconstruction of time series in state space. The conventional methods (false nearest neighbours and first minimum of the mutual information for estimating the embedding dimension and time lag, respectively) are also included for comparison purposes. The performance of GA and conventional parameters are tested by a one step ahead prediction modelling and estimation of dynamic invariants (i.e. approximate entropy). The results of this study indicated that the parameters selected by GA provide a better reconstruction (i.e. lower root mean square error) of EEG signals used for a Brain-Computer Interface (BCI) application. Additionally, GA based parameters are found to be computationally less intensive since both parameters are jointly optimised. In order to further illustrate the superiority of the embedding parameters estimated by GA, approximate entropy (ApEn) features using embedding parameters estimated by GA and conventional methods were computed. Next these ApEn features were used to classify the EEG signals into two classes (movement and non-movement) for BCI application. These results show that the embedding parameters estimated by GA are more appropriate than those estimated by the conventional methods for nonlinear modelling of EEG signals in state space.

We present a novel two module scheme for efficient analysis of noisy Electrocardiogram (ECG) signals. The first module consists of a segmentation algorithm which uses cyclostationary analysis for the detection of a single heart beat or cycle (P wave-QRS complex-T wave). The time domain cyclostationary (CS) algorithm exploits the statistical properties of the recorded periodic ECG signal and does not use any prior knowledge about signal morphology. Using the obtained cycle length the next module uses repeated applications of Principal Component Analysis (PCA) to reduce multiple additive noises from the multi trial and multi channel recorded ECG signals. PCA has been used for noise reduction in ECG but the method of repeated applications of PCA is novel. In this study, PCA was applied in 2 stages. In the first stage, PCA was applied to multi-channel ECG signals from one trial. The output ECG signals from the first stage were used in the second stage, where PCA was applied to multi-trial ECG signals from a single channel. The proposed scheme was tested with the 12-lead ECG signals from PTB Diagnostic database (National Metrology Institute of Germany) provided on physionet website which showed significant improvement in Signal to Noise ratio. We suggest that this simple scheme can be used for automatic analysis of noisy ECG signals where the extraction and denoising of single heart beat provide enhanced physiological features which enables better clinical interpretation of cardiovascular functionalities.

An enhanced online adaptive centre allocation algorithms (or resource allocation network (RAN)) using simple/stochastic back-propagation method with minimal weight update variant are developed for direct-link radial basis function (DRBF) networks. These algorithms are developed primarily for applications with fast sampling rate which demands significant reduction in computation load per iteration. The new algorithms are evaluated on a chaotic nonlinear biological based time series signals such as electroencephalographic (EEG) and electrocardiography (ECG). The EEG and ECG signals not only shows non-stationary behaviour but also large oscillation or changes. When the sample time is in milliseconds, both neural network adaptation and weight update must take place within the short time frame thus any learning rule must be computationally simple. The second order techniques, such as extended Kalman filter (EKF), need large amount of memory O(N2) and computationally intensive. The main goal of this paper is to develop a simple back-propagation based (SBP) resource allocation network (RAN), or also known as sequential learning technique using Radial Basis Function by incorporating Gaussian kernel, in order to identify (model) EEG and ECG signals. Simulation results show the modeled data show good representation of the original signals with less prediction error.

This paper proposes the use of exponential autoregressive (EAR) model for modelling of time series that are known to exhibit non-linear dynamics such as random fluctuations of amplitude and frequency. Biological signal (bio-signal) such as electroencephalogram (EEG) is known to exhibit nonlinear dynamics. Such signals cannot be modelled with traditional linear modelling techniques like autoregressive (AR) models as these models are known to provide only an approximation to the underlying properties of the non-linear signals. In this study, the suitability of EAR models as compared to AR models is shown using EEG signals in addition to several non-linear benchmark time series data where improved signal to noise ratio (SNR) values are indicated by the EAR models. Overall, the results indicate that use of EAR modelling which has yet to be exploited for bio-signal time series analysis has the huge potential in the characterisation and classification of EEG signals.

A method based on event-related brain signal is used for data encryption. The idea is to shuffle the Huffman tree using an encryption key generated by electroencephalogram (EEG) signals recorded when the user perceives a common black and white line picture. As different persons have different thought processes, the generated key is unique to each individual and hence the encryption is robust to fraudulent attacks as compared to other encryption systems. Further, as Huffman tree is used to encode the data during encryption, the method achieves both compression and encryption. This pilot study has shown the huge potential of the method as it is impossible to be compromised.

In this paper, we explore the use of genetic algorithm (GA) to select a minimum number of channels that identifies individuals based on brain signals i.e. electroencephalogram (EEG). The fusion of GA with linear discriminant classifier shows that the identification performance of EEG signals from 40 subjects does not degrade when using 23 selected channels as compared to all the available 61 channels as studied previously. As the channel identification method by GA is general, it could be used in any feature reduction application.

In this study, we investigated the linearity/nonlinearity of mental activity electroencephalogram (EEG) signals for Brain-Computer Interface (BCI) designs using the recent but well established Delay Vector Variance (DVV) method. EEG data recorded from seven subjects while they were performing five different mental activities were used in the experimental study. Through the use of DVV, it was investigated whether EEG signals would become linear or nonlinear when segmented into smaller parts. Concluding, the results of the studies showed that a large percentage of the EEG signals exhibited non-linear behaviour. This is an important result as it shows that the currently used linear modelling methods are mostly unsuitable.

A technique is proposed to reduce additive noise from biomedical signals that have high kurtosis values using genetic algorithm (GA). The technique is applied to reduce multiple linear additive noises from electrocardiogram (ECG) signals, which have high kurtosis values due to the presence of R peaks. This GA method uses the basic principles of Independent Component Analysis (ICA) and could also be used to reduce additive noise from other signals that have high kurtosis values. The method is simpler compared to neural learning algorithms and does not require any prior statistical knowledge of the signals. An additional advantage of the method compared to other ICA methods is that only the ECG signal will be extracted thus avoiding extraction of all independent components and manual inspection to determine the ECG signal.

Brain activity based biometric in the context of authentication of individuals is investigated. Specifically, a novel two-stage authenticating procedure to authenticate the claimed identity of the user using evoked potentials obtained when the user perceives a visual stimulus is proposed. A thorough analysis is performed on the proposed pilot approach, which validates the success of the approach.

A genetic algorithm is proposed for ordering the input patterns during training for Simplified Fuzzy ARTMAP (SFA) classifier to improve the individual identification classification performance using brain fingerprints. The results indicate improved classification performance as compared to the existing methods for pattern ordering, namely voting strategy and min-max. As the ordering method is general, it could be used with any dataset to obtain improved classification performance when SFA is used.

The dream of controlling and guiding computer-based systems using human brain signals has slowly but steadily become a reality. The available technology allows real-time implementation of systems that measure neuronal activity, convert their signals, and translate their output for the purpose of controlling mechanical and electronic systems. This paper describes the state of the art of non-invasive Brain-Machine Interfaces (BMIs) and critically investigates both the current technological limits and the future potential that BMIs have for space applications. We present an assessment of the advantages that BMIs can provide and justify the preferred candidate concepts for space applications together with a vision of future directions for their implementation.

In this paper, repeated applications of Principal Component Analysis (PCA) are proposed to reduce background electroencephalogram (EEG) artifact from multi-channel and multi-trial Visual Evoked Potential (VEP) signals. This will allow single trial analysis of VEP signals. PCA has been used for noise reduction but the method of repeated applications of PCA is novel. In the study here, PCA was applied in 2 stages. In the first stage, PCA was applied to multi-channel VEP signals from one trial. The output VEP signals from the first stage were used in the second stage, where PCA was applied to multi-trial VEP signals from a single channel. Simulation study using emulated VEP signals contaminated with EEG artifact shows significant improvement in signal to noise ratio using the method. It was then applied to study the electrophysiological differences between alcoholic and nonalcoholic subjects using N4 parameter. Hypothesis testing using t-test showed that alcoholics had significantly weaker and slower N4 responses as compared to non-alcoholics.

In recent years, numerous Brain Computer Interface (BCI) technologies have been developed to assist the disabled. In this paper, mental task based BCI is proposed for a different purpose: to identify the individuality of a person. The idea is based on the classification of electroencephalogram (EEG) signals recorded when a user thinks of either one or two mental tasks. As different individuals have different thought processes, this idea would be appropriate for individual identification. To increase the inter-subject differences, EEG data from six electrodes are used instead of one. Sixth order autoregressive features are computed from EEG signals and classified by Linear Discriminant classifier using a modified 10 fold cross validation procedure, which gave an average error of 0.95% when tested on 400 EEG patterns from four subjects. Though the method would have to undergo further development to obtain repeatable good accuracy; this initial study has shown the huge potential of the method over existing biometric identification systems as it is impossible to be faked.

In this paper, a Brain Computer Interface (BCI) is designed using electroencephalogram (EEG) signals where the subjects have to think of only a single mental task. The method uses spectral power and power difference in 4 bands: delta and theta, beta, alpha and gamma. This could be used as an alternative to the existing BCI designs that require classification of several mental tasks. In addition, an attempt is made to show that different subjects require different mental task for minimising the error in BCI output. In the experimental study, EEG signals were recorded from 4 subjects while they were thinking of 4 different mental tasks. Combinations of resting (baseline) state and another mental task are studied at a time for each subject. Spectral powers in the 4 bands from 6 channels are computed using the energy of the Elliptic FIR filter output. The mental tasks are detected by a neural network classifier. The results show that classification accuracy up to 97.5% is possible, provided that the most suitable mental task is used. As an application, the proposed method could be used to move a cursor on the screen. If cursor movement is used with a translation scheme like Morse Code, the subjects could use the proposed BCI for constructing letters/words. This would be very useful for paralysed individuals to communicate with their external surroundings.

In single trial source separation problem of VEP signals, the selection of legitimate Principal Components (PCs) is an important phenomenon. The Spectral Power Ratio (SPR) method developed by us earlier for PCA has proven to be capable of selecting only the required pes in a sophisticated manner. Our continuous enhancement has lead to the current development of the proposed method, Sandwich SPR (SSPR). The SSPR performs the reconstruction of source signal in an effective way better than the related SPR method. When this technique was applied on artificial Visual Evoked Potential (VEP) signals contaminated with background electroencephalogram (EEG), with a focus on extracting P3 parameters, it was found to be feasible shown by the resulting high values of the Signal to Noise ratio (SNR) as compared to the SPR and 2 tier SPR (SPR2) methods. Subsequently, we applied this method to study the P3 amplitude responses from a set of real EEG from Wadsworth BCI dataset obtained with target and non-target stimuli,d and found that the P3 parameters extracted through our proposed SSPR method showed higher P3 responses for the target stimuli than the both SPR and SPR2 methods, which conform to the existing knowledge on P3 responses.

The purpose of this study is to motivate the use of the simpler Linear Discriminant (LD) classifier as compared to the commonly used Multilayer-perceptron-backpropagation (MLP-BP) neural network for Brain Computer Interface (BCI) design. We investigated the performances of MLP-BP and LD classifiers for mental task based BCI design. In the experimental study, EEG signals from five mental tasks were recorded from four subjects and the classification performances of different combinations of two mental tasks were studied for each subject. Two different AR models were used to compute the features from the electroencephalogram signals: Burg's algorithm (ARB) and Least Square algorithm (ARLS). The results showed that in most cases, LD classifier gave superior classification performance as compared to MLP-BP, with reduced computational complexity. However, the best mental tasks for each subject were the same using both classifiers. ARLS gave the best performance (93.10%) using MLP-BP and (97.00%) using LD. As the best mental task combinations varied between subjects, we conclude that for different subjects, proper selection of mental tasks and feature extraction methods would be essential for a BCI design.

It has been shown previously that recognizing persons using 40 Hz electroencephalogram (EEC) oscillations is possible. In the method, features were computed from the Visual Evoked Potential (VEP) signals recorded from 61 electrodes while subjects perceived a picture. Here, two modifications have been proposed to improve the classification performance: Principal Component Analysis (PCA) to reduce the noise and background EEC effects from the VEP signals and normalization. Two classifiers were used: Simplified Fuzzy ARTMAP (SFA), and k-Ncarcst Neighbor (kNN). The experimental results using 800 VEP signals from 20 subjects with leavc-onc-out cross validation strategy showed that PCA and normalization improved the classification performance for both the classifiers. The best classification performance of 95.25% obtained using the improved method shows that 40 Hz EEC oscillations arc suitable for use as biometrics.

In this paper, a novel method to recognise persons using their brain patterns is proposed. These brain patterns are obtained when the individuals perceive a picture. High frequency brain energy is used as features that are classified by Elman backpropagation neural network. The experimental results using 1600 brain signals from 40 individuals give average classification rate of 96.63%. This pilot investigation shows that the proposed method of recognising persons using their brain signals is worth further study.

A segmentation algorithm, which detects a single cardiac cycle (S 1-Systole-S2-Diastole) of Phonocardiogram (PCG) signals using Homomorphic filtering and K-means clustering and a three way classification of heart sounds into Normal (N), Systolic murmur (S) and Diastolic murmur (D) using Grow and Learn (GAL) neural network, are presented. Homomorphic filtering converts a non-linear combination of signals (multiplied in time domain) into a linear combination by applying logarithmic transformation. It involves the retrieval of the envelope, a(n) of the PCG signal by attenuating the contribution of fast varying component, f(n) using an appropriate low pass filter. K-means clustering is a non-hierarchical partitioning method, which helps to indicate single cardiac cycle in the PCG signal. Segmentation performance of 90.45% was achieved using the proposed algorithm. Feature vectors were formed after segmentation by using Daubechies-2 wavelet detail coefficients at the second decomposition level. Grow and Learn network was used for classification of the segmented PCG signals and a classification accuracy of 97.02% was achieved. It is concluded that Homomorphic filtering and GAL network could be used for segmentation and classification of PCG signals without using a reference signal.

An algorithm for segmentation of heart sounds (HSs) into a single cardiac cycle (S 1-Systole-S 2-Diastole) using homomorphic filtering and k-means clustering and a three way classification of heart sounds into Normal (N), Systolic murmur (S), and Diastolic murmur (D), based on neural networks is developed. This algorithm does not require additional reference signal such as ECG signal. Feature vectors are formed after segmentation by using Daubechies-2 wavelet detail coefficients at the second decomposition level. Redundant features are removed using principal component analysis (PCA). Multilayer perceptron-Backpropagation neural network (MLP-BP) is used for classification of three different HSs. A classification accuracy of 94.5% and a segmentation accuracy (or performance) of 90.45% was achieved; thus, demonstrating that segmentation and classification of heart sounds without the aid of reference signal is achievable.

In this paper we propose a novel method to classify alcohol abusers. The method described efficiently estimated total power in gamma band spectral power (GBSP) of multi-channel visual evoked potential (VEP) signals in the time domain, circumventing power spectrum computation. Then, the total power extracted are used as features to classify alcohol abusers from control subjects using Multilayer Perceptron -Back Propagation (MLP-BP) neural network classifier. As a comparison study the total power using GBSP feature extraction is repeated for four types of Infinite Impluse Response (IIR) filters. Experimental study is conducted with 20 subjects totaling 800 VEP signals, which are extracted while subjects are seeing pictures from Snodgrass and V anderwart set. Maximum classification of 91% is obtained for Elliptic filter for 20 hidden units. Also Elliptic filter shows the best performance for the averaged values of all the filters and it also has the lower order when compared to other filters.

We investigate the decision making ability of subjects with a history of long-term intoxicant (alcohol) consumption in relation to short-term visual memory using P3 amplitudes obtained from single trial visual evoked potential (VEP) signals. This is made possible by means of digital filtering and principal component analysis (PCA). The results show a significantly lower P3 amplitude for these subjects as compared to controls. This electroencephalogram based analysis conforms with the common knowledge that long-term alcohol use causes permanent and negative residual effects on decision making ability related to short-term visual memory.

In recent years, numerous Brain Computer Interface (BCI) technologies have been developed to assist the disabled. In this paper, mental task based BCI is proposed for a different purpose: to identify the individuality of a person. The idea is based on the classification of electroencephalogram (EEG) signals recorded when a user thinks of either one or two mental tasks. As different individuals have different thought processes, this idea would be appropriate for individual identification. To increase the inter-subject differences, EEG data from six electrodes are used instead of one. Sixth order autoregressive features are computed from EEG signals and classified by Linear Discriminant classifier using a modified 10 fold cross validation procedure, which gave an average error of 0.95% when tested on 400 EEG patterns from four subjects. Though the method would have to undergo further development to obtain repeatable good accuracy; this initial study has shown the huge potential of the method over existing biometric identification systems as it is impossible to be faked.

In this paper, we propose using a combination of ECG and blood pressure signals to detect ectopic heart beats, specifically premature supraventricular and ventricular contractions (PSC and PVC). Detection of these beats are important are they could be pre-cursor for serious arrhythmias. Common detection methods of these beats use only ECG signals. However, the stroke volume changes after the occurrence of these beats, which results in blood pressure variations. Following this fact, we combined features extracted from QRS complex of ECG signal Lead I with systolic and diastolic arterial blood pressure values to classify normal, PSC and PVC beats. Data from 5 subjects totaling 750 beats (250 normal, 250 PSC and 250 PVC) from Massachusetts General Hospital/Marquette Foundation (MGH/MF) database were used. The data were split equally for Multilayer Perceptron - Backpropagation (MLP-BP) neural network training and testing. The combined features were classified by the MLP-BP neural network into the 3 classes. The features were normalized using some parameter inherent in the signals. This was to normalize the features across different subjects. The results gave classification performance up to 92.00%. It is concluded that ECG and blood pressure features could detect PSC and PVC.

Classification of EEG signals extracted during mental tasks is a technique for designing Brain Computer Interfaces (BCI). In this paper, we classify EEG signals that were extracted during mental tasks using fixed autoregressive (FAR) and adaptive AR (AAR) models. Five different mental tasks from 4 subjects were used in the experimental study and combinations of 2 different mental tasks are studied for each subject. Four different feature extraction methods were used to extract features from these EEG signals: FAR coefficients computed with Burg's algorithm using 125 data points, without segmentation and with segmentation of 25 data points, AAR coefficients computed with Least-Mean-Square (LMS) algorithm using 125 data points, without segmentation and with segmentation of 25 data points. Multilayer Perceptron (MLP) neural network (NN) trained by the backpropagation (BP) algorithm is used to classify these features into the different categories representing the mental tasks. The best results for FAR was 92.70% while for AAR was only 81.80%. The results obtained here indicated that FAR using 125 data points without segmentation gave better classification performance as compared to AAR, with all other parameters constant.

In this paper, we propose a technique to identify individuals using features extracted from QRS segment of electrocardiogram (ECG) signals. A total of 2000 samples from 10 subjects from the Arrhythmia Laboratory at Boston's Beth Israel Hospital (now the Beth Israel Deaconess Medical Center) database were used. These data are available as MIT-BH Normal Sinus Rhythm database and consist of 18 hour long-term recordings with 2 ECG signals. The commonly used features like R-R interval, R amplitude, QRS interval, QR amplitude and RS amplitude were used. In addition to these features, we propose the use of form factor of the QRS segment. Form factor has been used previously in electroencephalogram analysis and it is a measure of the complexity of the signal. These six features were then used by two neural network classifiers: Multilayer Perceptron - Backpropagation (MLP-BP) and Simplified Fuzzy ARTMAP (SFA). The data were split equally for MLP-BP and SFA training and testing. The results gave classification performance up to 97.6%. This indicates that ECG has the potential to be used as a biometric tool.

We analyse features from R waveform of electrocardiogram (ECG) and blood pressure (BP) signals for intelligent detection of ectopic (premature) heart beats. Detection of these beats are important are they could be pre-cursor for serious arrhythmias. The combination of ECG and blood pressure signals to detect ectopic heart beats are relatively new as most methods use only ECG signals. Two new features, mobility and complexity factor (used in electroencephalogram analysis) derived from R waveform of ECG signal are studied in addition to the conventional ECG and BP features. Three cases of beats are considered: 2 ectopic - premature ventricular contraction (PVC) and premature supraventricular contraction (PSC) and normal (N). All the features are normalized with some factor inherent in the signal to reduce the inter-subject variance of the features. Data from 50 subjects totaling 3000 beats (1000 N, 1000 PSC and 1000 PVC) from Massachusetts General Hospital/Marquette Foundation database are used. The 13 features were classified by the Multilayer Perceptron - Backpropagation neural network into the 3 classes. The results gave classification performance up to 96.47%. It is concluded that ECG and BP features could be used to detect ectopic beats successfully.

A new method to identify individuals using signals from the brain is proposed In the method, brain signals specifically known as Visual Evoked Potential (VEP) signals are recorded from 61 electrodes located on the scalp while the individuals see a picture. Next, spectral features are computed and classified by a Simplified Fuzzy ARTMAP (SFA) neural network (NN). The spectral features consist of power in gamma band (30-50 Hz). The experimental results using 800 VEP signals from 20 subjects show give average classification of 94.18%. This pilot investigation shows that the proposed method of identifying individuals using their brain signals is worth further study.

We investigate the relationship of N1-P2 evoked response (peak-to-peak amplitude and time interval) with short-term visual memory in humans. Visual evoked responses obtained from 20 subjects (10 non-amnesic alcoholics and 10 non-alcoholics) are extracted from channel P8 referenced to channel Cz during the presentation of modified delayed matching-to-sample visual task using Snodgrass and Vanderwart picture set. Our results indicate that N1-P2 amplitudes are higher for non-matching (novel) stimuli as compared to matching stimuli for all the subjects. N1-P2 time interval is also shorter for the case of matching stimuli. This indicates that information processing is increased for the non-matching stimuli as compared to matching stimuli. These results are quite consistent with a number of related studies and we conclude that N1-P2 is related to short-term visual memory involved during object recognition. The results also indicate that N1-P2 amplitude is higher for non-alcoholics as compared to alcoholics, which indicates that some form of memory impairment exist in alcoholics.

Two common artifacts that corrupt evoked responses are noise and background electroencephalogram (EEG). In this paper, a two-level principal component analysis (PCA) is used to reduce these artifacts from single trial evoked responses. The first level PCA is applied to reduce noise from these VEP signals while the second level PCA reduces EEG. The method is used to analyse the object recognition and decision-making capability during visual responses. The analysis is extended to study the differences in visual response between alcoholics and non-alcoholics using single trial P3 visual evoked potential (VEP) signals. The analysis shows that alcoholics respond slower and weaker to visual stimulus as compared to non-alcoholics.

Proposes a technique that uses genetic algorithm (GA) to select optimal features for classification applications using fuzzy ARTMAP (FA) neural network (NN). The technique is applied to select features for classification of two groups of subjects: alcoholics and controls, using multi-channel single trial electroencephalogram (EEG) signals evoked during visual response. The results show that the proposed technique is successful in selecting the features that contribute towards classification. This serves to reduce the number of required features while improving the classification performance. The results also indicate that the gamma band spectral power could be used to support evidence on the residual effects of long-term use of alcohol on visual response.

We describe a method to extract single trial Visual Evoked Potential (VEP) buried in ongoing Electroencephalogram (EEG) activity. The common method for separating VEP from EEG is to use signal averaging. But we use digital filters to extract VEP assuming that VEP spectra are in the gamma band. As an application, a Fuzzy ARTMAP (FA) neural network classifier with voting strategy is used with this extracted VEP to discriminate alcoholics from normal subjects. The VEP is extracted from subjects while seeing visuals of Snodgrass and Vanderwart picture set. The high FA classification of 96.5% shows the validity of the proposed method to successfully remove EEG contamination.

We use the power and asymmetry ratio of spectral bands to recognise mental tasks from electroencephalogram signals using a fuzzy ARTMAP neural network. Classical spectral analysis using the Wiener-Khintchine theorem and modem parametric spectral analysis using the autoregressive method are used to obtain these features. The highest classification results of 90% for a subject recognising two mental tasks validate the method.

A new method of communication is proposed for paralysed patients using EEG signals. A scheme based on Morse code is used to construct meaningful words using recognised mental tasks. A benefit of this system is as a means of communication between paralysed patients and their external environment i.e. as an interface for use by people with severe physical disabilities. As the technology advances, it is envisaged that this technique could be used by anyone for rudimentary user-interface actions, like popping up windows and making menu choices. The EEG signals are segmented and power spectral values are extracted using Wiener-Khintchine theorem with Parzen window smoothing. A Fuzzy ARTMAP (FA) classifier is used to classify these signals into three different mental tasks, where each task either represents dit, dah or space. We have analysed different mental tasks and show that the performance with different tasks varies greatly for different subjects. The FA classification results in this paper have shown that it is highly possible to recognise three different mental tasks for each subject provided that these three tasks, which varies for different subjects are chosen after some preliminary simulations.

Visual inspection of EEG signals in their unprocessed form is still the predominant way of discriminating EEG patterns in the medical community and requires highly trained medical professionals. To overcome this problem, automatic EEG analysis using Fourier Transform methods are popular since most EEG signals consist of spectral power in the range of δ, θ, α and β i.e from 0 to 30 Hz. But this method suffers from high noise sensitivity and is not suitable for short and variable length of signal segments. In this paper, we analyze EEG signals with time series analysis using autoregression techniques. We classify these extracted features for different mental tasks using a Fuzzy ARTMAP classifier. We study the effects of different EEG segment or window lengths and different overlapping lengths on the overall performance of the classifier. Our results show that the segment length affects the performance and that overlapping the segments improves the performance greatly.

A new technique of fusing genetic algorithms with Fuzzy ARTMAP is proposed. This method selects the appropriate autoregressive model order for EEG signals and consequently classifies these signals into their respective different mental tasks. The experimental results show that this method outperforms other statistical autoregressive model order selection methods like Akaike Information Criterion, Final Prediction Error and reflection, coefficient.

Regular moment invariants face two limitations. First, images with symmetry in the x and/or y directions and symmetry at centroid give zero values for odd orders of central moments. Secondly, they are very sensitive to noise, especially the higher order moments. This paper presents a single solution to solve the symmetrical problem and reduce the noise sensitivity of these moments. The solution involves a new set of moment-based features that uses a reference point other than the image centroid. The reference centre is selected such that the new moment features are invariant to translation, scaling and rotation. The derivation of the new moments and their invariance are shown before experimenting them with some symmetrical alphabets. Next, they are shown to be less sensitive under the presence of Gaussian and random noise as compared to the usual regular moment invariants. Noise corrupted English alphabets are classified with a neural network to further verify the advantage of using the new moment features.