Electroencephalogram (EEG) signal analysis is one of the mostly studied research areas in biomedical signal processing, and machine learning. Emotion recognition through machine intelligence plays ...critical role in understanding the brain activities as well as in developing decision-making systems. In this research, an automated EEG based emotion recognition method with a novel fractal pattern feature extraction approach is presented. The presented fractal pattern is inspired by Firat University Logo and named fractal Firat pattern (FFP). By using FFP and Tunable Q-factor Wavelet Transform (TQWT) signal decomposition technique, a multilevel feature generator is presented. In the feature selection phase, an improved iterative selector is utilized. The shallow classifiers have been considered to denote the success of the presented TQWT and FFP based feature generation. This model has been tested on emotional EEG signals with 14 channels using linear discriminant (LDA), k-nearest neighborhood (k-NN), support vector machine (SVM). The proposed framework achieved 99.82% with SVM classifier.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Deep learning (DL)-based brain–computer interface (BCI) in motor imagery (MI) has emerged as a powerful method for establishing direct communication between the brain and external electronic devices. ...However, due to inter-subject variability, inherent complex properties, and low signal-to-noise ratio (SNR) in electroencephalogram (EEG) signals are major challenges that significantly hinder the accuracy of the MI classifier.
To overcome this, the present work proposes an efficient transfer learning (TL)-based multi-scale feature fused CNN (MSFFCNN) which can capture the distinguishable features of various non-overlapping canonical frequency bands of EEG signals from different convolutional scales for multi-class MI classification.
In order to account for inter-subject variability from different subjects, the current work presents 4 different model variants including subject-independent and subject-adaptive classification models considering different adaptation configurations to exploit the full learning capacity of the classifier. Each adaptation configuration has been fine-tuned in an extensively trained pre-trained model and the performance of the classifier has been studied for a vast range of learning rates and degrees of adaptation which illustrates the advantages of using an adaptive transfer learning-based model.
The model achieves an average classification accuracy of 94.06% (±0.70%) and the kappa value of 0.88 outperforming several baseline and current state-of-the-art EEG-based MI classification models with fewer training samples. The present research provides an effective and efficient transfer learning-based end-to-end MI classification framework for designing a high-performance robust MI-BCI system.
•An efficient TL-based MSFFCNN has been proposed that demonstrates superior performance for EEG-based MI classification in BCIs.•The MS-CNN block enhances the overall model performance from multiple scales.•The OVR-FBCSP CNN block can efficiently extract the discriminative spatiotemporal CSP features of ERD/ERS.•Proposed model achieves average classification accuracy of 94.06% outperforming several SOAT models with fewer training samples.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Electroencephalogram (EEG) signals contain vital information on the electrical activities of the brain and are widely used to aid epilepsy analysis. A challenging element of epilepsy diagnosis, ...accurate classification of different epileptic states, is of particular interest and has been extensively investigated. A new deep learning-based classification methodology, namely epileptic EEG signal classification (EESC), is proposed in this paper. This methodology first transforms epileptic EEG signals to power spectrum density energy diagrams (PSDEDs), then applies deep convolutional neural networks (DCNNs) and transfer learning to automatically extract features from the PSDED, and finally classifies four categories of epileptic states (interictal, preictal duration to 30 min, preictal duration to 10 min, and seizure). It outperforms the existing epilepsy classification methods in terms of accuracy and efficiency. For instance, it achieves an average classification accuracy of over 90% in a case study with CHB-MIT epileptic EEG data.
Brain activities can be evaluated by using Electroencephalogram (EEG) signals. One of the primary reasons for traffic accidents is driver fatigue, which can be identified by using EEG signals. This ...work aims to achieve a highly accurate and straightforward process to detect driving fatigue by using EEG signals. Two main problems, which are feature generation and feature selection, are defined to achieve this aim. This work solves these problems by using two different approaches. Deep networks are efficient feature generators and extract features in low, medium, and high levels. These features can be generated by using multileveled or multilayered feature extraction. Therefore, we proposed a multileveled feature generator that uses a one-dimensional binary pattern (BP) and statistical features together, and levels are created using a one-dimensional discrete wavelet transform (1D-DWT). A five-level fused feature extractor is presented by using BP, statistical features of 1D-DWT together. Moreover, a 2-layered feature selection method is proposed using ReliefF and iterative neighborhood component analysis (RFINCA) to solve the feature selection problem. The goals of the RFINCA are to choose the optimal number of features automatically and use the effectiveness of ReliefF and neighborhood component analysis (NCA) together. A driving fatigue EEG dataset was used as a testbed to denote the effectiveness of eighteen conventional classifiers. According to the experimental results, a highly accurate EEG classification approach is presented. The proposed method also reached 100.0% classification accuracy by using a k-nearest neighborhood classifier.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK
Background: Psychiatrists diagnose schizophrenia based on clinical symptoms such as disordered thinking, delusions, hallucinations, and severe distortion of daily functions. However, some of these ...symptoms are common with other mental illnesses such as bipolar mood disorder. Therefore, quantitative assessment of schizophrenia by analyzing a physiological-based data such as the electroencephalogram (EEG) signal is of interest. In this study, we analyze the spectrum and time-frequency distribution (TFD) of EEG signals to understand how schizophrenia affects these signals. Methods: In this regard, EEG signals of 20 patients with schizophrenia and 20 age-matched participants (control group) were investigated. Several features including spectral flux, spectral flatness, spectral entropy, time-frequency (TF)-flux, TF-flatness, and TF-entropy were extracted from the EEG signals. Results: Spectral flux (1.5388±0.0038 and 1.5497±0.0058 for the control and case groups, respectively, P=0.0000), spectral entropy (0.8526±0.0386 and 0.9018±0.0428 for the control and case groups, respectively, P=0.0004), spectral roll-off (0.3896±0.0434 and 0.4245±0.0410 for the control and case groups, respectively, P=0.0129), spectral flatness (0.1401±0.0063 and 0.1467±0.0077 for the control and case groups, respectively, P=0.0055), TF-flux (1.2675±0.1806 and 1.5284±0.2057 for the control and case groups, respectively, P=0.0001) and TF-flatness (0.9980±0.0000 and 0.9981±0.0000 for the control and case groups, respectively, P=0.0000) values in patients with schizophrenia were significantly greater than the control group in most EEG channels. This prominent irregularity may be caused by decreasing the synchronization of neurons in the frontal lobe. Conclusion: Spectral and time frequency distribution analysis of EEG signals can be used as quantitative indexes for neurodynamic investigation in schizophrenia.
The EEG signal classification is crucial for epileptic seizure prediction. Therefore, many machine learning model has been presented to classify EEG signals accurately.
This work presents a novel ...automated EEG classification method by using a novel nonlinear feature extractor, and it is called as Hamsi-Pat. It uses the substitution box (S-Box) of the Hamsi hash function. As stated in the literature, S-Boxes have generally used for diffusion in symmetric encryption (especially block ciphers) methods and cryptologic hash functions. Since it is a nonlinear structure, this work aims to illustrate the merit of an S-Box for feature generation. Therefore, a new generation feature generator, which is Hamsi-Pat, is presented by using S-Box of the Hamsi hash function, and a novel EEG classification method is proposed by using Hamsi-Pat. The presented biomedical signal classification method has three elementary phases, and these phases are Hamsi-Pat based multileveled feature generation, iterative neighborhood component analysis (INCA) selector based feature dimension reduction, and classification using k nearest neighborhood (kNN) classifier. The presented Hamsi-Pat and INCA based methods were tested on Bonn electroencephalography (EEG) datasets.
This model yielded 99.20% classification accuracy on the used EEG dataset for five classes case and it yielded 100.0% accuracies for other cases.
These results obviously denoted that the S-Boxes can be considered as a feature generator, and a novel S-Box based feature generation research area can be defined as textural feature generation and statistical feature generation.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Electroencephalography (EEG) signals recorded during mental arithmetic tasks can be used to quantify mental performance. The classification of these input EEG signals can be automated using machine ...learning models. We aimed to develop an efficient handcrafted model that could accurately discriminate “bad counters” vs. “good counters” in mental arithmetic.
We studied a public mental arithmetic task performance EEG dataset comprising 20-channel EEG signal segments recorded from 36 healthy right-handed subjects divided into two classes 10 “bad counters” and 26 “good counters”. The original 60-second EEG samples are divided into 424 15-second segments (119 and 305 belonging to the “bad counters” and “good counters”, respectively) to input into our model. Our model comprised a novel multilevel feature extraction method based on (1) four rhombuses lattice pattern, a new generation function for feature extraction that was inspired by the lattice structure in post-quantum cryptography; and (2) triangle pooling, a new distance-based pooling function for signal decomposition. These were combined with downstream feature selection using iterative neighborhood component analysis, channel-wise result classification using support vector machine with leave-one-subject-out (LOSO) and 10-fold) cross-validations (CVs) to calculate prediction vectors, iterative majority voting to generate voted vectors, and greedy algorithm to obtain the best results.
The model attained 88.44% and 96.42% geometric means and accuracies of 93.40% and 97.88%, using LOSO and 10-fold CVs, respectively.
Our model’s >93% classification accuracies compared favorably against published literature. Importantly, the model has linear computational complexity, which enhances its ease of implementation.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Nowadays, many deep models have been presented to recognize emotions using electroencephalogram (EEG) signals. These deep models are computationally intensive, it takes a longer time to train the ...model. Also, it is difficult to achieve high classification performance using for emotion classification using machine learning techniques. To overcome these limitations, we present a hand-crafted conventional EEG emotion classification network. In this work, we have used novel prime pattern and tunable q-factor wavelet transform (TQWT) techniques to develop an automated model to classify human emotions. Our proposed cognitive model comprises feature extraction, feature selection, and classification steps. We have used TQWT on the EEG signals to obtain the sub-bands. The prime pattern and statistical feature generator are employed on the generated sub-bands and original signal to generate 798 features. 399 (half of them) out of 798 features are selected using minimum redundancy maximum relevance (mRMR) selector, and misclassification rates of each signal are evaluated using support vector machine (SVM) classifier. The proposed network generated 87 feature vectors hence, this model is named PrimePatNet87. In the last step of the feature generation, the best 20 feature vectors which are selected based on the calculated misclassification rates, are concatenated. The generated feature vector is subjected to the feature selection and the most significant 1000 features are selected using the mRMR selector. These selected features are then classified using an SVM classifier. In the last phase, iterative majority voting has been used to generate a general result. We have used three publicly available datasets, namely DEAP, DREAMER, and GAMEEMO, to develop our proposed model. Our presented PrimePatNet87 model reached over 99% classification accuracy on whole datasets with leave one subject out (LOSO) validation. Our results demonstrate that the developed prime pattern network is accurate and ready for real-world applications.
•Automated detection of emotion recognition using EEG signals.•Novel prime pattern network is proposed is proposed to generate the features.•Proposed method is developed using three public databases.•Obtained an accuracy of more than 99% for three databases.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
•A novel method using 1D-LBPs and 1-NN classifier for classification of seizure and seizure-free EEG signals.•Bank of Gabor filters for decomposing EEG signal.•Three different schemes have been ...investigated for fusion of information.•Approach achieves the best 10-fold classification accuracy of 98.33%.
Local binary pattern (LBP) is a texture descriptor that has been proven to be quite effective for various image analysis tasks in image processing. In this paper one-dimensional local binary pattern (1D-LBP) based features are used for classification of seizure and seizure-free electroencephalogram (EEG) signals. The proposed method employs a bank of Gabor filters for processing the EEG signals. The processed EEG signal is divided into smaller segments and histograms of 1D-LBPs of these segments are computed. Nearest neighbor classifier utilizes the histogram matching scores to determine whether the acquired EEG signal belongs to seizure or seizure-free category. Experimental results on publicly available database suggest that the proposed features effectively characterize local variations and are useful for classification of seizure and seizure-free EEG signals with a classification accuracy of 98.33%. This result demonstrates the superiority of our approach for classification of seizure and seizure-free EEG signals over recently proposed approaches in the literature.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK
This paper describes the analysis of a deep neural network for the classification of epileptic EEG signals. The deep learning architecture is made up of two convolutional layers for feature ...extraction and three fully-connected layers for classification. We evaluated several EEG signal transforms for generating the inputs to the deep neural network: Fourier, wavelet and empirical mode decomposition. This analysis was carried out using two public datasets (Bern-Barcelona EEG and Epileptic Seizure Recognition datasets) obtaining significant improvements in accuracy. For the Bern-Barcelona EEG, we obtained an increase in accuracy from 92.3% to 98.9% when classifying between focal and non-focal signals using the empirical mode decomposition. For the Epileptic Seizure Recognition dataset, we evaluated several scenarios for seizure detection obtaining the best results when using the Fourier transform. The accuracy increased from 99.0% to 99.5% for classifying non-seizure vs. seizure recordings, from 91.7% to 96.5% when differentiating between healthy, non-focal and seizure recordings, and from 89.0% to 95.7% when considering healthy, focal and seizure recordings.
•This paper analyzes the use of a deep neural network for epileptic EEG signal classification.•The deep architecture is composed of 2 convolutional layers for features extraction and 3 fully connected layers for classification.•Analysis of different inputs to the DNN: raw data, Fourier transform, wavelet coefficients and empirical mode decomposition.•EMD is interesting for classifying focal vs. non-focal signals and Fourier transform performs better for seizure detection.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP