Design and implementation of an SVM-based computer classification system for discriminating depressive patients from healthy controls using the P600 component of ERP signals

A comparative analysis of masking empirical mode decomposition and a neural network with feed-forward and back propagation along with masking empirical mode decomposition to improve the classification performance for a reliable brain-computer interface

In general, extraction and classification are used in various fields like image processing, pattern recognition, signal processing, and so on. Extracting effective characteristics from raw electroencephalogram (EEG) signals is a crucial role of the brain-computer interface for motor imagery. Recently, there has been a great deal of focus on motor imagery in the EEG signals since they encode a person's intent to do an action. Researchers have been using MI signals to assist paralyzed people and even move them on their own with certain equipment, like wheelchairs. As a result, proper decoding is an important step required for the interconnection of the brain and the computer. EEG decoding is a challenging process because of poor SNR, complexity, and other reasons. However, choosing an appropriate method to extract the features to improve the performance of motor imagery recognition is still a research hotspot. To extract the features of the EEG signal in the classification task, this paper proposes a Masking Empirical Mode Decomposition (MEMD) based Feed Forward Back Propagation Neural Network (MEMD-FFBPNN). The dataset consists of EEG signals which are first normalized using the minimax method and given as input to the MEMD to extract the features and then given to the FFBPNN to classify the tasks. The accuracy of the proposed method MEMD-FFBPNN has been measured using the confusion matrix, mean square error and which has been recorded up to 99.9%. Thus, the proposed method gives better accuracy than the other conventional methods.

Machine Learning Techniques for the Diagnosis of Schizophrenia Based on Event-Related Potentials

Antecedent

The event-related potential (ERP) components P300 and mismatch negativity (MMN) have been linked to cognitive deficits in patients with schizophrenia. The diagnosis of schizophrenia could be improved by applying machine learning procedures to these objective neurophysiological biomarkers. Several studies have attempted to achieve this goal, but no study has examined Multiple Kernel Learning (MKL) classifiers. This algorithm finds optimally a combination of kernel functions, integrating them in a meaningful manner, and thus could improve diagnosis.

Objective

This study aimed to examine the efficacy of the MKL classifier and the Boruta feature selection method for schizophrenia patients (SZ) and healthy controls (HC) single-subject classification.

Methods

A cohort of 54 SZ and 54 HC participants were studied. Three sets of features related to ERP signals were calculated as follows: peak related features, peak to peak related features, and signal related features. The Boruta algorithm was used to evaluate the impact of feature selection on classification performance. An MKL algorithm was applied to address schizophrenia detection.

Results

A classification accuracy of 83% using the whole dataset, and 86% after applying Boruta feature selection was obtained. The variables that contributed most to the classification were mainly related to the latency and amplitude of the auditory P300 paradigm.

Conclusion

This study showed that MKL can be useful in distinguishing between schizophrenic patients and controls when using ERP measures. Moreover, the use of the Boruta algorithm provides an improvement in classification accuracy and computational cost.

Adding Tactile Feedback and Changing ISI to Improve BCI Systems' Robustness: An Error-Related Potential Study

Nowadays, the brain-computer interface (BCI) systems attract much more attention than before, yet they have not found their ways into our lives since their accuracy is not satisfying. Error Related Potential (ErRP) is a potential that occurs in human brain signals when an unintended event happens, against ones' will and thoughts. An example is the occurrence of an error in BCI systems. Investigation of the ErRP could enable researchers to increase the accuracy of BCI systems by detecting instances of inaccuracy in the system. In this research the effects of two parameters on the ErRP are studied: (1) The Motor Imagery Time, also known as Inter-Stimulus Interval (ISI) and (2) different types of feedback (Visual and Tactile). The statistical analysis of the ErRP characteristics showed that feedback type meaningfully affects the ErRP in a cue-paced BCI system and it will affect the time of occurrence of this potential. To validate the proposed idea, different feature extraction, and classification techniques were used for the classification of the BCI system responses. It was shown that by proper selection of the parameters and features, the accuracy of the system could be improved. Tactile feedback together with higher ISI could increase the accuracy of finding erroneous trials up to 90%. The proposed method's accuracy was significantly higher (p-value < 0.05) compared to other methods of feature extraction.

Electrical brain activity and facial electromyography responses to irony in dysphoric and non-dysphoric participants

We studied irony comprehension and emotional reactions to irony in dysphoric and control participants. Electroencephalography (EEG) and facial electromyography (EMG) were measured when spoken conversations were presented with pictures that provided either congruent (non-ironic) or incongruent (ironic) contexts. In a separate session, participants evaluated the congruency and valence of the stimuli. While both groups rated ironic stimuli funnier than non-ironic stimuli, the control group rated all the stimuli funnier than the dysphoric group. N400-like activity, P600, and EMG activity indicating smiling were larger after the ironic stimuli than the non-ironic stimuli for both groups. Further, in the dysphoric group the irony modulation was evident in the electrode cluster over the right hemisphere, while no such difference in lateralization was observed in the control group. The results suggest a depression-related alteration in the P600 response associated to irony comprehension, but no alterations were found in emotional reactivity specifically related to irony.

Classification of Depression Through Resting-State Electroencephalogram as a Novel Practice in Psychiatry: Review

Background

Machine learning applications in health care have increased considerably in the recent past, and this review focuses on an important application in psychiatry related to the detection of depression. Since the advent of computational psychiatry, research based on functional magnetic resonance imaging has yielded remarkable results, but these tools tend to be too expensive for everyday clinical use.

Objective

This review focuses on an affordable data-driven approach based on electroencephalographic recordings. Web-based applications via public or private cloud-based platforms would be a logical next step. We aim to compare several different approaches to the detection of depression from electroencephalographic recordings using various features and machine learning models.

Methods

To detect depression, we reviewed published detection studies based on resting-state electroencephalogram with final machine learning, and to predict therapy outcomes, we reviewed a set of interventional studies using some form of stimulation in their methodology.

Results

We reviewed 14 detection studies and 12 interventional studies published between 2008 and 2019. As direct comparison was not possible due to the large diversity of theoretical approaches and methods used, we compared them based on the steps in analysis and accuracies yielded. In addition, we compared possible drawbacks in terms of sample size, feature extraction, feature selection, classification, internal and external validation, and possible unwarranted optimism and reproducibility. In addition, we suggested desirable practices to avoid misinterpretation of results and optimism.

Conclusions

This review shows the need for larger data sets and more systematic procedures to improve the use of the solution for clinical diagnostics. Therefore, regulation of the pipeline and standard requirements for methodology used should become mandatory to increase the reliability and accuracy of the complete methodology for it to be translated to modern psychiatry.

Attentional bias in MDD: ERP components analysis and classification using a dot-probe task

BACKGROUND AND OBJECTIVE

Strands of evidence have supported existence of negative attentional bias in patients with depression. This study aimed to assess the behavioral and electrophysiological signatures of attentional bias in major depressive disorder (MDD) and explore whether ERP components contain valuable information for discriminating between MDD patients and healthy controls (HCs).

METHODS

Electroencephalography data were collected from 17 patients with MDD and 17 HCs in a dot-probe task, with emotional-neutral pairs as experimental materials. Fourteen features related to ERP waveform shape were generated. Then, Correlated Feature Selection (CFS), ReliefF and GainRatio (GR) were applied for feature selection. For discriminating between MDDs and HCs, k-nearest neighbor (KNN), C4.5, Sequential Minimal Optimization (SMO) and Logistic Regression (LR) were used.

RESULTS

Behaviorally, MDD patients showed significantly shorter reaction time (RT) to valid than invalid sad trials, with significantly higher bias score for sad-neutral pairs. Analysis of split-half reliability in RT indices indicated a strong reliability in RT, while coefficients of RT bias scores neared zero. These behavioral effects were supported by ERP results. MDD patients had higher P300 amplitude with the probe replacing a sad face than a neutral face, indicating difficult attention disengagement from negative emotional faces. Meanwhile, data mining analysis based on ERP components suggested that CFS was the best feature selection algorithm. Especially for the P300 induced by valid sad trials, the classification accuracy of CFS combination with any classifier was above 85%, and the KNN (k = 3) classifier achieved the highest accuracy (94%).

CONCLUSIONS

MDD patients show difficulty in attention disengagement from negative stimuli, reflected by P300. The CFS over other methods leads to a good overall performance in most cases, especially when KNN classifier is used for P300 component classification, illustrating that ERP component may be applied as a tool for auxiliary diagnosis of depression.

Classifying Schizotypy Using an Audiovisual Emotion Perception Test and Scalp Electroencephalography

Schizotypy refers to the personality trait of experiencing "psychotic" symptoms and can be regarded as a predisposition of schizophrenia-spectrum psychopathology (Raine, 1991). Cumulative evidence has revealed that individuals with schizotypy, as well as schizophrenia patients, have emotional processing deficits. In the present study, we investigated multimodal emotion perception in schizotypy and implemented the machine learning technique to find out whether a schizotypy group (ST) is distinguishable from a control group (NC), using electroencephalogram (EEG) signals. Forty-five subjects (30 ST and 15 NC) were divided into two groups based on their scores on a Schizotypal Personality Questionnaire. All participants performed an audiovisual emotion perception test while EEG was recorded. After the preprocessing stage, the discriminatory features were extracted using a mean subsampling technique. For an accurate estimation of covariance matrices, the shrinkage linear discriminant algorithm was used. The classification attained over 98% accuracy and zero rate of false-positive results. This method may have important clinical implications in discriminating those among the general population who have a subtle risk for schizotypy, requiring intervention in advance.

A Computer-Aided Diagnosis System With EEG Based on the P3b Wave During an Auditory Odd-Ball Task in Schizophrenia

OBJECTIVE

To design a Computer-aided diagnosis (CAD) system using an optimized methodology over the P3b wave in order to objectively and accurately discriminate between healthy controls (HC) and schizophrenic subjects (SZ).

METHODS

We train, test, analyze, and compare various machine learning classification approaches optimized in terms of the correct classification rate (CCR), the degenerated Youden's index (DYI) and the area under the receiver operating curve (AUC). CAD system comprises five stages: electroencephalography (EEG) preprocessing, feature extraction, seven electrode groupings, discriminant feature selection, and binary classification.

RESULTS

With two optimal combinations of electrode grouping, filtering, feature selection algorithm, and classification machine, we get either a mean CCR = 93.42%, specificity = 0.9673, sensitivity = 0.8727, DYI = 0.9188, and AUC = 0.9567 (total-15 Hz-J5-MLP), or a mean CCR = 92.23%, specificity = 0.9499, sensitivity = 0.8838, DYI = 0.9162, and AUC = 0.9807 (right hemisphere-35 Hz-J5-SVM), which to our knowledge are higher than those available to date.

CONCLUSIONS

We have verified that a more restrictive low-pass filtering achieves higher CCR as compared to others at higher frequencies in the P3b wave. In addition, results validate previous hypothesis about the importance of the parietal-temporal region, associated with memory processing, allowing us to identify powerful {feature,electrode} pairs in the diagnosis of schizophrenia, achieving higher CCR and AUC in classification of both right and left Hemispheres, and parietal-temporal EEG signals, like, for instance, the {PSE, P4} pair (J5 and mutual information feature selection).

SIGNIFICANCE

Diagnosis of schizophrenia is made thoroughly by psychiatrists but as any human-based decision that has a subjective component. This CAD system provides the human expert with an objective complimentary measure to help him in diagnosing schizophrenia.

Reprint of "A new approach to analyze data from EEG-based concealed face recognition system"

The purpose of this study is to extend a feature set with non-linear features to improve classification rate of guilty and innocent subjects. Non-linear features can provide extra information about phase space. The Event-Related Potential (ERP) signals were recorded from 49 subjects who participated in concealed face recognition test. For feature extraction, at first, several morphological characteristics, frequency bands, and wavelet coefficients (we call them basic-features) are extracted from each single-trial ERP. Recurrence Quantification Analysis (RQA) measures are then computed as non-linear features from each single-trial. We apply Genetic Algorithm (GA) to select the best feature set and this feature set is used for classification of data using Linear Discriminant Analysis (LDA) classifier. Next, we use a new approach to improve classification results based on introducing an adaptive-threshold. Results indicate that our method is able to correctly detect 91.83% of subjects (45 correct detection of 49 subjects) using combination of basic and non-linear features, that is higher than 87.75% for basic and 79.59% for non-linear features. This shows that combination of non-linear and basic- features could improve classification rate.

A new approach to analyze data from EEG-based concealed face recognition system

The purpose of this study is to extend a feature set with non-linear features to improve classification rate of guilty and innocent subjects. Non-linear features can provide extra information about phase space. The Event-Related Potential (ERP) signals were recorded from 49 subjects who participated in concealed face recognition test. For feature extraction, at first, several morphological characteristics, frequency bands, and wavelet coefficients (we call them basic-features) are extracted from each single-trial ERP. Recurrence Quantification Analysis (RQA) measures are then computed as non-linear features from each single-trial. We apply Genetic Algorithm (GA) to select the best feature set and this feature set is used for classification of data using Linear Discriminant Analysis (LDA) classifier. Next, we use a new approach to improve classification results based on introducing an adaptive-threshold. Results indicate that our method is able to correctly detect 91.83% of subjects (45 correct detection of 49 subjects) using combination of basic and non-linear features, that is higher than 87.75% for basic and 79.59% for non-linear features. This shows that combination of non-linear and basic- features could improve classification rate.

EEG classification of adolescents with type I and type II of bipolar disorder

Bipolar disorder (BD) is a severe psychiatric disorder and has two common types: type I and type II. Early diagnosis of the subtypes is very challenging particularly in adolescence. In this study, 38 adolescents are participated including 18 patients with BD I and 20 patients with BD II. The electroencephalogram signal is recorded by 19 electrodes in open eyes at resting state. After preprocessing, the state of the art methods from various domains are implemented to provide a good feature set for classifying the two groups. In order to improve the classification accuracy, four different feature selection methods named mutual information maximization (MIM), conditional mutual information maximization (CMIM), fast correlation based filter (FCBF), and double input symmetrical relevance (DISR) are applied to select the most informative features. Multilayer perceptron (MLP) neural network with a hidden layer containing five neurons is used for classification with and without applying the feature selection methods. The accuracy of 82.68, 86.33, 89.67, 84.61, and 91.83 % were observed using entire extracted features and selected features using MIM, CMIM, FCBF, and DISR methods by MLP, respectively. Therefore, the proposed method can be used in clinical setting for more validation.

Neural Correlates of Craving in Methamphetamine Abuse

Introduction:

Methamphetamine is a powerful psychostimulant that causes significant neurological impairments with long-lasting effects and has provoked serious international concerns about public health. Denial of drug abuse and drug craving are two important factors that make the diagnosis and treatment extremely challenging. Here, we present a novel and rapid noninvasive method with potential application for differentiation and monitoring methamphetamine abuse.

Methods:

Visual stimuli comprised a series of images with neutral and methamphetamine-related content. A total of 10 methamphetamine abusers and 10 age-gender matched controls participated in the experiments. Event-related potentials (ERPs) were recorded and compared using a time window analysis method. The ERPs were divided into 19 time windows of 100 ms with 50 ms overlaps. The area of positive sections below each window was calculated to measure the differences between the two groups.

Results:

Significant differences between two groups were observed from 250 to 500 ms (P300) in response to methamphetamine-related visual stimuli and 600 to 800 ms in response to neutral stimuli.

Conclusion:

This study presented a novel and noninvasive method based on neural correlates to discriminate healthy individuals from methamphetamine drug abusers. This method can be employed in treatment and monitoring of the methamphetamine abuse.

Real-time seizure prediction using RLS filtering and interpolated histogram feature based on hybrid optimization algorithm of Bayesian classifier and Hunting search

BACKGROUND AND OBJECTIVES

Epileptic seizure prediction using EEG signal analysis is an important application for drug therapy and pediatric patient monitoring. Time series estimation to obtain the future samples of EEG signal has vital role for detecting seizure attack. In this paper, a novel density-based real-time seizure prediction algorithm based on a trained offline seizure detection algorithm is proposed.

METHODS

In the offline seizure detection procedure, after signal preprocessing, histogram-based statistical features are extracted from signal probability distribution. By defining a deterministic polynomial model on the normalized histogram, a novel syntactic feature that is named Interpolated Histogram Feature (IHF) is proposed. Moreover, with this feature, Seizure Distribution Model (SDM) as a descriptor of the seizure and non-seizure signals is presented. By using a novel hybrid optimization algorithm based on Bayesian classifier and Hunting Search (HuS) algorithm, the optimal features are selected. To detect the seizure attacks in the online mode, a Multi-Layer Perceptron (MLP) classifier is trained with the optimal features in the offline procedure. For online prediction, the enhanced Recursive Least Square (RLS) filter is applied to estimate sample-by-sample of the EEG signal. Also, a density-based signal tracking scenario is introduced to update and tune the parameters of RLS filtering algorithm.

RESULTS

Our prediction algorithm is evaluated on 104 hours of EEG signals recorded from 23 pediatric patients. Our online signal prediction algorithm provides the accuracy rate of 86.56% and precision rate of 86.53% simultaneously using the trained MLP classifier from the offline mode. The recall rate of seizure prediction is 97.27% and the false prediction rate of 0.00215 per hour is achieved as well. Ultimately, the future samples of EEG signal are estimated, and the time of seizure signal prediction is also converged to 6.64 seconds.

CONCLUSION

In our proposed real-time algorithm, by implementing a density-based signal tracking scenario, the future samples of signal with suitable time is predicted and the seizure is detected based on the optimal features from the IHF and histogram-based statistical features with acceptable performance.

Advanced analysis of auditory evoked potentials in hyperthyroid patients: the effect of filtering

The purpose of the study is to evaluate Auditory Evoked Potentials (AEPs) in patients with hyperthyroidism and to compare their frequency components with those of healthy subjects. In this study the AEPs in hyperthyroidism were studied both in time and frequency domains rather than studying just in the time domain by peak scoring. This paper presents a method for filtering auditory oddball standard and target AEPs by using singular spectrum analysis (SSA) and feature extraction in the frequency domain via spectral analysis. AEPs were recorded during an auditory oddball paradigm in 25 newly diagnosed hyperthyroid patients and 15 healthy subjects. The signals are captured in the presence of ongoing background EEG activity so they are often contaminated by artifacts. This paper presents a method for filtering auditory odd-ball standard and target AEPs by using Singular spectrum analysis and feature extraction in frequency domain via spectral analysis. Information about the frequency composition of the signal is then used to compare normal and hyperthyroid states. While there was no significant difference either in the target or standard unfiltered signals between the hyperthyroid patients and the control group (p > 0.05), there was a significant difference in the filtered signals between the two groups (p < 0.01). In conclusion, our results revealed that SSA is an effective filtering method for AEPs. Thus, a much more objective and specific examination method was developed.

A novel algorithm to enhance P300 in single trials: application to lie detection using F-score and SVM

The investigation of lie detection methods based on P300 potentials has drawn much interest in recent years. We presented a novel algorithm to enhance signal-to-noise ratio (SNR) of P300 and applied it in lie detection to increase the classification accuracy. Thirty-four subjects were divided randomly into guilty and innocent groups, and the EEG signals on 14 electrodes were recorded. A novel spatial denoising algorithm (SDA) was proposed to reconstruct the P300 with a high SNR based on independent component analysis. The differences between the proposed method and our/other early published methods mainly lie in the extraction and feature selection method of P300. Three groups of features were extracted from the denoised waves; then, the optimal features were selected by the F-score method. Selected feature samples were finally fed into three classical classifiers to make a performance comparison. The optimal parameter values in the SDA and the classifiers were tuned using a grid-searching training procedure with cross-validation. The support vector machine (SVM) approach was adopted to combine with an F-score because this approach had the best performance. The presented model F-score_SVM reaches a significantly higher classification accuracy for P300 (specificity of 96.05%) and non-P300 (sensitivity of 96.11%) compared with the results obtained without using SDA and compared with the results obtained by other classification models. Moreover, a higher individual diagnosis rate can be obtained compared with previous methods, and the presented method requires only a small number of stimuli in the real testing application.

Identifying Differences in the P600 Component of ERP-Signals between OCD Patients and Controls Employing a PNN-based Majority Vote Classification Scheme

In the present study an attempt was made to focus in the differences between Obsessive-Compulsive Disorder (OCD) patients and healthy controls, as reflected by the P600 component of event-related potential (ERP) signals, to locate brain areas that may be related to Working Memory (WM) deficits. Neuropsychological research has yielded contradicting results regarding WM in OCD. Eighteen patients with OCD symptomatology and 20 normal controls (age and sex matched) were subjected to a computerized version of the digit span Wechsler test. EEG activity was recorded from 15 scalp electrodes (leads). A dedicated computer software was developed to read the ERP signals and to calculate features related to the ERP P600 component (500-800 ms). Nineteen features were generated, from each ERP-signal and each lead, and were employed in the design of the Probabilistic Neural Network (PNN) classifier. Highest single-lead precision (86.8%) was found at the Fp2 and C6 leads. When the output from all single-lead PNN classifiers fed a Majority Vote Engine (MVE), the system classified correctly all subjects, providing a powerful classification scheme. Findings indicated that OCD patients differed from normal controls at the prefrontal and temporo-central brain regions.

Effect of latency on clustering of P300 recordings for ADHD discrimination

This paper is focused on testing the latency contribution as regards the quality of formed groups for discriminating between healthy and attention deficit hyperactivity disorder children. To this end, two different cases are considered: nonaligned original recordings and aligned signals according to P300 position. For latter case, a novel approach to conduct time location of P300 component is introduced, which is based on derivative of event-related potential signals. The used database holds event-related potentials registered in auditory and visual oddball paradigm. Several experiments are carried out testing both configurations of considered data matrix. For grouping input data matrices, the k-means clustering technique is employed. To assess the quality of formed clusters and the relevance for clustering of latency-based features, relative values of distances between centroids and data points are computed in order to apprise separability and compactness of estimated clusters. Experimental results show that time localization of P300 component is not a decisive feature in formation of compact and well-defined groups within a discrimination framework for two considered data classes under certain conditions.

A comparison of different feature extraction methods for diagnosis of valvular heart diseases using PCG signals

This article presents a novel method for diagnosis of valvular heart disease (VHD) based on phonocardiography (PCG) signals. Application of the pattern classification and feature selection and reduction methods in analysing normal and pathological heart sound was investigated. After signal preprocessing using independent component analysis (ICA), 32 features are extracted. Those include carefully selected linear and nonlinear time domain, wavelet and entropy features. By examining different feature selection and feature reduction methods such as principal component analysis (PCA), genetic algorithms (GA), genetic programming (GP) and generalized discriminant analysis (GDA), the four most informative features are extracted. Furthermore, support vector machines (SVM) and neural network classifiers are compared for diagnosis of pathological heart sounds. Three valvular heart diseases are considered: aortic stenosis (AS), mitral stenosis (MS) and mitral regurgitation (MR). An overall accuracy of 99.47% was achieved by proposed algorithm.

Support vector machine classification of multi-channel EEG traces: a new tool to analyze the brain response to morphine treatment

The analgesic effect of morphine is highly individual, calling for objective methods to predict the subjective pain relief. Such methods might be based on alteration of brain response caused by morphine during painful stimuli. The study included 11 healthy volunteers subjectively quantifying perception of painful electrical stimulations in the esophagus. Brain evoked potentials following stimulations were recorded from sixty-four electroencephalographic channels at baseline and ninety minutes after morphine administration. Marginals obtained from discrete wavelet coefficients for each channel were used as input to an optimized support vector machine classifying between baseline and after morphine administration. The electroencephalographic channel leading to the best performance was further analyzed to identify brain alterations caused by morphine. Marginals from volunteers with no analgesic effect were examined for differences in comparison to volunteers with effect. The single-channel classification showed best performance at electrode P4 with 84.1 % of the traces classified correctly. When combining features from the 6 best performing channels, the multichannel classification increased to 92.4 %. The most discriminative feature was a decrease in the delta band (0.5 - 4 Hz) after morphine for volunteers with analgesic effect. Volunteers with no effect of morphine showed an increase in the delta band after drug administration. As only a proportion of patients benefit from opioid treatment, the new approach may help to identify non-responders and guide individualized tailored analgesic therapy.

Classification of event-related potentials associated with response errors in actors and observers based on autoregressive modeling

Event-Related Potentials (ERPs) provide non-invasive measurements of the electrical activity on the scalp related to the processing of stimuli and preparation of responses by the brain. In this paper an ERP-signal classification method is proposed for discriminating between ERPs of correct and incorrect responses of actors and of observers seeing an actor making such responses. The classification method targeted signals containing error-related negativity (ERN) and error positivity (Pe) components, which are typically associated with error processing in the human brain. Feature extraction consisted of Multivariate Autoregressive modeling combined with the Simulated Annealing technique. The resulting information was subsequently classified by means of an Artificial Neural Network (ANN) using back-propagation algorithm under the “leave-one-out cross-validation” scenario and the Fuzzy C-Means (FCM) algorithm. The ANN consisted of a multi-layer perceptron (MLP). The approach yielded classification rates of up to 85%, both for the actors’ correct and incorrect responses and the corresponding ERPs of the observers. The electrodes needed for such classifications were situated mainly at central and frontal areas. Results provide indications that the classification of the ERN is achievable. Furthermore, the availability of the Pe signals, in addition to the ERN, improves the classification, and this is more pronounced for observers’ signals. The proposed ERP-signal classification method provides a promising tool to study error detection and observational-learning mechanisms in performance monitoring and joint-action research, in both healthy and patient populations.

A new approach for EEG feature extraction in P300-based lie detection

P300-based Guilty Knowledge Test (GKT) has been suggested as an alternative approach for conventional polygraphy. The purpose of this study was to extend a previously introduced pattern recognition method for the ERP assessment in this application. This extension was done by the further extending the feature set and also the employing a method for the selection of optimal features. For the evaluation of the method, several subjects went through the designed GKT paradigm and their respective brain signals were recorded. Next, a P300 detection approach based on some features and a statistical classifier was implemented. The optimal feature set was selected using a genetic algorithm from a primary feature set including some morphological, frequency and wavelet features and was used for the classification of the data. The rates of correct detection in guilty and innocent subjects were 86%, which was better than other previously used methods.

Improving accuracy in astrocytomas grading by integrating a robust least squares mapping driven support vector machine classifier into a two level grade classification scheme

Grading of astrocytomas is an important task for treatment planning; however, it suffers from significantly great inter-observer variability. Computer-assisted diagnosis systems have been propose to assist towards minimizing subjectivity, however, these systems present either moderate accuracy or utilize specialized staining protocols and grading systems that are difficult to apply in daily clinical practice. The present study proposes a robust mathematical formulation by integrating state-of-art technologies (support vector machines and least squares mapping) in a cascade classification scheme for separating low from high and grade III from grade IV astrocytic tumours. Results have indicated that low from high-grade tumours can be correctly separated with a certainty as high as 97.3%, whereas grade III from grade IV tumours with 97.8%. The overall performance was 95.2%. These high rates have been a result of applying the least squares mapping technique to features prior to classification. A significant byproduct of least squares mapping is that the number of support vectors of the SVM classifiers dropped dramatically from about 80% when no mapping was used to less than 5% when mapping was used. The latter is a clear indication that the SVM classifier has a greater potential to generalize well to new data. In this way, digital image analysis systems for automated grading of astrocytomas are brought closer to clinical practice.