You are on page 1of 2

1-)An approach to estimate EEG power spectrum as an index of heat stress using backpropagation artificial

neural network
A method has been presented for an effective application of backpropagation artificial neural network (ANN) in establishment of electroencephalogram (EEG) power spectra as an index of stress in hot environment. The power spectrum data for slow wave sleep (SWS), rapid eye movement (REM) sleep and awake (AWA) states in three groups of rats (acute heat stress, chronic heat stress and the normal) were tested by an ANN, containing 60 nodes in input layer, weighted from power spectrum data from 0 to 30 Hz, 18 nodes in hidden layer and an output node. The target output values for this network were determined with another fivelayered neural network (with the structure of 3-12-1-12-3). The input and output of this network was assigned with the three wellestablished heat stress indices (body temperature, body weight and plasma corticosterone). The most important feature for acute stress, chronic stress and normal conditions were extracted from the third layer single neuron and used for the target value for the three-layered neural network. The ANN was found effective in recognising the EEG power spectra with an average of 96.67% for acute heat stress, 97.17% for chronic heat stress and 98.5% for normal subjects.

2-)Automatic epileptic seizure detection in EEGs based on line length feature and artificial neural Networks
About 1% of the people in the world suffer from epilepsy. The main characteristic of epilepsy is the recurrent seizures. Careful analysis of the electroencephalogram (EEG) recordings can provide valuable information for understanding the mechanisms behind epileptic disorders. Since epileptic seizures ocur irregularly and unpredictably, automatic seizure detection in EEG recordings is highly required. Wavelet transform (WT) is an effective analysis tool for non-stationary signals, such as EEGs. The line length feature reflects the waveform dimensionality changes and is a measure sensitive to variation of the signal amplitude and frequency. This paper presents a novel method for automatic epileptic seizure detection, which uses line length features based on wavelet transform multiresolution decomposition and combines with an artificial neural network (ANN) to classify the EEG signals regarding the existence of seizure or not. To the knowledge of the authors, there exists no similar work in the literature. A famous public dataset was used to evaluate the proposed method. The high accuracy obtained for three different classification problems testified the great success of the method.

3-)The ANN-based computing of drowsy level


We have developed a new method for automatic estimation of vigilance level by using electroencephalogram (EEG), electromyogram (EMG) and eye movement (EOG) signals recorded during transition from wakefulness to sleep. In the previous studies, EEG signals and EEG signals with EMG signals were used for estimating vigilance levels. In the present study, it was aimed to estimate vigilance levels by using EEG, EMG and EOG signals. The changes in EEG, EMG and EOG were diagnosed while transiting from wakefulness to sleep by using wavelet transform and developed artificial neural network (ANN). EEG signals were separated to its subbands using wavelet transform, LEOG (Left EOG), REOG (Right EOG) and chin EMG was used in ANN process for increasing the accuracy of the estimation rate by evaluating their tonic levels and also used in data preparation stage to verify and eliminate the movement artifacts. Then, training and testing data sets consist of the EEG subbands (delta, theta, alpha and beta); LEOG, REOG and EMG signals were applied to the ANN for training and testing the system which gives three situations for the vigilance level of the subject: Awake, drowsy, and sleep. The accuracy of estimation is about 9798% while the accuracy of the previous study which used only EEG was 9596% and the study which used EEG with EMG was 9899%. The reason of decreasing the percentage of present study according to the last study is because of the increase of the input data.

4-) Adaptive training using an artificial neural network and EEG metrics for within- and cross-task workload classification
Adaptive training using neurophysiological measures requires efficient classification of mental workload in real time as a learner encounters new and increasingly difficult levels of tasks. Previous investigations have shown that artificial neural networks (ANNs) can accurately classify workload, but only when trained on neurophysiological exemplars from experienced operators on specific tasks. The present study examined classification accuracies for ANNs trained on electroencephalographic (EEG) activity recorded while participants performed the same (within task) and different (cross) tasks for short periods of time with little or no prior exposure to the tasks. Participants performed three working memory tasks at two difficulty levels with order of task and difficulty level counterbalanced. Within-task classification accuracies were high when ANNs were trained on exemplars from the same task or a set containing the to-be-classified task, (M=87.1% and 85.3%, respectively). Cross-task classification accuracies were significantly lower (average 44.8%) indicating consistent systematic misclassification for certain tasks in some individuals. Results are discussed in terms of their implications for developing neurophysiologically driven adaptive training platforms. 5-) Recognition of mental activity-dependent EEG patterns using artificial neural network This study was devoted to the investigation of human mental activity during spatial and verbal task solving. We aimed to classify different task types using artificial neural network (ANN) as a recognition system, based on the analysis of EEG rhythms. Using the ANN allows good recognition after a single EEG trial because the ANN performs a parallel weighted estimation of many classification signs simultaneously. The recognition thus becomes possible in nearly real time, i.e. with only a few (about 23) seconds delay.

The best results were obtained using the smoothed Fourier power spectra of single EEG trials in 520 Hz frequency band as an input to ANN, i.e. when spatial and frequency properties of EEG rhythms form the base of classification. 480 tasks of eight different types belonging to two basic types (spatial and verbal) were presented in a random order to 20 healthy adults aged 2230. First, the ANN is programmed to recognize the task type correctly using the learning set of patterns, i.e. power spectra of single EEG trials corresponding to periods of task solving. At this stage, a type of mental task that corresponds to each EEG trial is assumed to be known. Later, the network recognizes the unknown (new) data. Thus, the recognition is based on the analysis of EEG rhythmical properties performed by ANN. Our results show that we can distinguish between principal mental activity types with confidence. If ANN is able to distinguish between verbal and spatial thinking with four types of tasks only (e.g. match, cube section, anagram, ranging) the other four (fragments, road, two words, contradiction) are recognized correctly as belonging to a proper basic class of mental activity with a high percent of correct recognition (85 10%, averaged over 20 subjects). This result demonstrates that the signs of mental task types really do exist and are well expressed in rhythmic properties of single EEG trials.