No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Classification of ADHD by Using Multiple Feature Maps of EEG Signals and Deep Feature Extraction
Emotion recognition using electroencephalogram (EEG) signals is getting more and more attention in recent years. Since the EEG signals are noisy, non-linear and have non-stationary properties, it is a challenging task to develop an intelligent framework that can provide high accuracy for emotion recognition. In this paper, we propose a new model for emotion recognition that will be based on the creation of feature maps based on the topographic (TOPO-FM) and holographic (HOLO-FM) representation of EEG signal characteristics. Deep learning has been utilized as a feature extractor method on feature maps, and afterward extracted features are fused together for the classification process to recognize different kinds of emotions. The experiments are conducted on the four publicly available emotion datasets: DEAP, SEED, DREAMER, and AMIGOS. We demonstrated the effectiveness of our approaches in comparison with studies where authors used EEG signals that classify human emotions in the two-dimensional space. Experimental results show that the proposed methods can improve the emotion recognition rate on the different size datasets.
Background
and objectives: Attention deficit hyperactivity disorder (ADHD) is often presented with conduct disorder (CD). There is currently no objective laboratory test or diagnostic method to discern between ADHD and CD, and diagnosis is further made difficult as ADHD is a common neuro-developmental disorder often presenting with other co-morbid difficulties; and in particular with conduct disorder which has a high degree of associated behavioural challenges.. A novel automated system (AS) is proposed as a convenient supplementary tool to support clinicians in their diagnostic decisions. To the best of our knowledge, we are the first group to develop an automated classification system to classify ADHD, CD and ADHD+CD classes using brain signals.
Methods
: The empirical mode decomposition (EMD) and discrete wavelet transform (DWT) methods were employed to decompose the electroencephalogram (EEG) signals. Autoregressive modelling coefficients and relative wavelet energy were then computed on the signals. Various nonlinear features were extracted from the decomposed coefficients. Adaptive synthetic sampling (ADASYN) was then employed to balance the dataset. The significant features were selected using sequential forward selection method. The highly discriminatory features were subsequently fed to an array of classifiers.
Results
: The highest accuracy of 97.88% was achieved with the K-Nearest Neighbour (KNN) classifier. The proposed system was developed using ten-fold validation strategy on EEG data from 123 children. To the best of our knowledge this is the first study to develop an AS for the classification of ADHD, CD and ADHD+CD classes using EEG signals.
Potential application
: Our AS can potentially be used as a web-based application with cloud system to aid the clinical diagnosis of ADHD and/or CD, thus supporting faster and accurate treatment for the children. It is important to note that testing with larger data is required before the AS can be employed for clinical applications.
The aim of this study was to build a machine learning model to discriminate Attention Deficit Hyperactivity Disorder (ADHD) patients and healthy controls using information from both time and frequency analysis of Event Related Potentials (ERP) obtained from Electroencephalography (EEG) signals while participants performed an auditory oddball task. The study included 23 unmedicated ADHD patients and 23 healthy controls. The EEG signal was analyzed in time domain by nonlinear brain dynamics and morphological features, and in time-frequency domain with wavelet coefficients. Selected features were applied to various machine learning techniques including; Multilayer Perceptron, Naïve Bayes, Support Vector Machines, k-nearest neighbor, Adaptive Boosting, Logistic Regression and Random Forest to classify ADHD patients and healthy controls. Longer P300 latencies and smaller P300 amplitudes were observed in ADHD patients relative to controls. In fractal dimension calculation relative to the control group, the ADHD group demonstrated reduced complexity. In addition, certain wavelet coefficients provided significantly different values in both groups. Combining these extracted features, our results indicated that Multilayer Perceptron method provided the best classification with an accuracy rate of 91.3% and a high level of reliability of concurrence (Kappa = 0.82).
The results showed that combining time and frequency domain features can be a useful and discriminative for diagnostic purposes in ADHD. The study presents a supporting diagnostic tool that uses EEG signal processing and machine learning algorithms. The findings would be helpful in the objective diagnosis of ADHD.
Attention Deficit Hyperactivity Disorder (ADHD) is a common behavioral disorder may be found in 5% to 8% of the children. Early diagnosis of ADHD is crucial for treating the disease and reducing its harmful effects on education, employment, relationships and life quality. On the other hand, nonlinear analysis methods are widely applied in processing
the Electroencephalogram (EEG) signals. It has been proved that the brain neuronal activity and its related EEG signals have chaotic behavior. Hence, chaotic indices can be employed to classify the EEG signals. In this study, a new approach is proposed based on the combination of some nonlinear features to distinguish ADHD from normal children. Lyapunov exponent, fractal dimension, correlation dimension and sample, fuzzy and approximate entropies are the nonlinear extracted features. For computing, the chaotic time series of obtained EEG in the brain frontal lobe (FP1, FP2, F3, F4 and Fz) need to be
analyzed. Experiments on a set of EEG signal obtained from 50 ADHD and 26 normal cases yielded a sensitivity, specificity and accuracy of 98%, 92.31%, and 96.05%, respectively. The obtained accuracy provides a significant improvement in comparison to the other similar studies in identifying and classifying children with ADHD.
The aim of this study is to recognize human emotions by electroencephalographic (EEG) signals. The innovation of our research methods involves two aspects: First, we integrate the spatial characteristics, frequency domain, and temporal characteristics of the EEG signals, and map them to a two-dimensional image. With these images, we build a series of EEG Multidimensional Feature Image (EEG MFI) sequences to represent the emotion variation with EEG signals. Second, we construct a hybrid deep neural network to deal with the EEG MFI sequences to recognize human emotional states where the hybrid deep neural network combined the Convolution Neural Networks (CNN) and Long Short-Term-Memory (LSTM) Recurrent Neural Networks (RNN). Empirical research is carried out with the open-source dataset DEAP (a Dataset for Emotion Analysis using EEG, Physiological, and video signals) using our method, and the results demonstrate the significant improvements over current state-of-the-art approaches in this field. The average emotion classification accuracy of each subject with CLRNN (the hybrid neural networks that we proposed in this study) is 75.21%.
Attention deficit hyperactivity disorder is a neurodevelopmental condition associated with varying levels of hyperactivity, inattention, and impulsivity. This study investigates brain function in children with attention deficit hyperactivity disorder using measures of nonlinear dynamics in EEG signals during rest. During eyes-closed resting, 19 channel EEG signals were recorded from 12 ADHD and 12 normal age-matched children. We used the multifractal singularity spectrum, the largest Lyapunov exponent, and approximate entropy to quantify the chaotic nonlinear dynamics of these EEG signals. As confirmed by Wilcoxon rank sum test, largest Lyapunov exponent over left frontal-central cortex exhibited a significant difference between ADHD and the age-matched control groups. Further, mean approximate entropy was significantly lower in ADHD subjects in prefrontal cortex. The singularity spectrum was also considerably altered in ADHD compared to control children. Evaluation of these features was performed by two classifiers: a Support Vector Machine and a Radial Basis Function Neural Network. For better comparison, subject classification based on frequency band power was assessed using the same types of classifiers. Nonlinear features provided better discrimination between ADHD and control than band power features. Under four-fold cross validation testing, support vector machine gave 83.33% accurate classification results.
This study investigates the non-linear features of electroencephalogram signals regarding ADHD and normal adult participants while performing Continuous Performance Test. Three non-linear features were extracted from the EEG signals. ADHD and age-matched normal groups were investigated separately which revealed that there is a significant relation between clinical presentation of the participants and some non-linear features. The accuracy of 88% and 96% were achieved in classification of clinical and non-clinical participants using one and two features respectively. The best classification result was obtained with a combination of two features in Wavelet-Entropy group.
A multi-paradigm methodology is presented for electroencephalogram (EEG) based diagnosis of Attention-Deficit/Hyperactivity Disorder (ADHD) through adroit integration of nonlinear science; wavelets, a signal processing technique; and neural networks, a pattern recognition technique. The selected nonlinear features are generalized synchronizations known as synchronization likelihoods (SL), both among all electrodes and among electrode pairs. The methodology consists of three parts: first detecting the more synchronized loci (group 1) and loci with more discriminative deficit connections (group 2). Using SLs among all electrodes, discriminative SLs in certain sub-bands are extracted. In part two, SLs are computed, not among all electrodes, but between loci of group 1 and loci of group 2 in all sub-bands and the band-limited EEG. This part leads to more accurate detection of deficit connections, and not just deficit areas, but more discriminative SLs in sub-bands with finer resolutions. In part three, a classification technique, radial basis function neural network, is used to distinguish ADHD from normal subjects. The methodology was applied to EEG data obtained from 47 ADHD and 7 control individuals with eyes closed. The Radial Basis Function (RBF) neural network classifier yielded a high accuracy of 95.6% for diagnosis of the ADHD in the feature space discovered in this research with a variance of 0.7%.
Abnormal functional brain connectivity is a candidate factor in developmental brain disorders associated with cognitive dysfunction.
We analyzed a substantial (10 min per subject) record of dense array electroencephalography with spectral power and coherence
methods in attention-deficit hyperactivity disorder (ADHD) (n = 42) and control (n = 21) 10- to 13-year-old children. We found topographically distinct narrow band coherence differences between subject groups:
ADHD subjects showed elevated coherence in the lower alpha (8 Hz) band and reduced coherence in the upper alpha (10–11 Hz)
band. The 8-Hz ADHD elevation and a 2- to 6-Hz control group coherence elevation were independent of stimulus presentation.
In response to visual stimulation, the ADHD group exhibited reduced evoked potential power and elevated frontal coherence.
Only the upper alpha band control group coherence elevation discriminated according to ADHD group medication status. The findings
suggest a static state of deficient connectivity in ADHD and a stimulus-induced state of overconnectivity within and between
frontal hemispheres.
Dementia is one of the most common neurological disorders causing defection of cognitive functions, and seriously affects the quality of life. In this study, various methods have been proposed for the detection and follow-up of Alzheimer's dementia (AD) with advanced signal processing methods by using electroencephalography (EEG) signals. Signal decomposition-based approaches such as empirical mode decomposition (EMD), ensemble EMD (EEMD), and discrete wavelet transform (DWT) are presented to classify EEG segments of control subjects (CSs) and AD patients. Intrinsic mode functions (IMFs) are obtained from the signals using the EMD and EEMD methods, and the IMFs showing the most significant differences between the two groups are selected by applying previously suggested selection procedures. Five-time-domain and 5-spectral-domain features are calculated using selected IMFs, and five detail and approximation coefficients of DWT. Signal decomposition processes are conducted for both 1 min and 5 s EEG segment durations. For the 1 min segment duration, all the proposed approaches yield prominent classification performances. While the highest classification accuracies are obtained using EMD (91.8%) and EEMD (94.1%) approaches from the temporal/right brain cluster, the highest classification accuracy for the DWT (95.2%) approach is obtained from the temporal/left brain cluster for 1 min segment duration.
Similarity quantification is an important field of study in electroencephalogram (EEG)-based brain activity detection, in which the goal is to compute interdependence between certain cortical areas from inter-hemispheric or intra-hemispheric channel pairs. This study aims to propose a new interdependence EEG feature, namely Dynamic frequency warpping(DFW) based on dynamic analysis of frequency fluctuations as a hybrid feature extraction step. A new EEG classifier based on sparse coding has been developed for Attention Deficit Hyperactivity Disorder (ADHD) detection. It has been tested using EEG recordings of 14 ADHD children and 19 healthy controls during resting state and a time-reproduction task. The capability of the proposed method with an accuracy rate of 99.17% has been shown. Use of the DFW extracted from frontal channel pairs or beta frequency band not only improves the performance but also reduces the computational complexity due to the need to a subgroup of channels or a subband.
Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental behavioral disorder. It is common in children, can be carried over into adulthood, and is associated with inattention, hyperactivity, and impulsive behavior. Physicians typically use the patient's description and questionnaires to diagnose this disorder. Due to its subjective nature, this procedure can lead to false diagnoses, which may cause irreparable distress to the patient's life. Since mental disorders are dependent on the brain function, researchers use biological signals such as electroencephalography (EEG) to help diagnose ADHD.
In this study, we propose a new feature extraction scheme based on evaluating dynamic connectivity tensors among EEG channels for constructing the input formulation of the classification model. The tensors contain correlations among the EEG channels over different time frames. This method allows preserving both temporal and spatial structures of the EEG data while reducing the input dimensions of the model. We then employ a neural network model consisting of a convolutional long short-term memory (ConvLSTM) and an attention mechanism to classify ADHD patients and the control group. This model can encode the spatiotemporal representation of EEG recordings and identify dependencies between temporal segments. Convolution is responsible for encoding and finding spatial dependencies between electrodes. LSTM explores the relationships between different time blocks, and finally, attention focuses on the most relevant parts of the sequence. We evaluate the proposed approach by performing experiments on the EEG dataset, including 400 instances with 30 s length collected from 46 children with ADHD and 45 children in the control group. We achieved an average accuracy of 99.34% on this dataset, and our best model has an accuracy of 99.75%, both are the highest among the work done in this field.
Detecting objects in images is an extremely important step in many image and video analysis applications. Object detection is considered as one of the main challenges in the field of computer vision, which focuses on identifying and locating objects of different classes in an image. In this paper, we aim to highlight the important role of deep learning and convolutional neural networks in particular in the object detection task. We analyze and focus on the various state-of-the-art convolutional neural networks serving as a backbone in object detection models. We test and evaluate them in the common datasets and benchmarks up-to-date. We Also outline the main features of each architecture. We demonstrate that the application of some convolutional neural network architectures has yielded very promising state-of-the-art results in image classification in the first place and then in the object detection task. The results have surpassed all the traditional methods, and in some cases, outperformed the human being’s performance.
The electroencephalogram (EEG) is an informative neuroimaging tool for studying attention-deficit/hyperactivity disorder (ADHD); one main goal is to characterize the EEG of children with ADHD. In this study, we employed the power spectrum, complexity and bicoherence, biomarker candidates for identifying ADHD children in a machine learning approach, to characterize resting-state EEG (rsEEG). We built support vector machine classifiers using a single type of feature, all features from a method (relative spectral power, spectral power ratio, complexity or bicoherence), or all features from all four methods. We evaluated effectiveness and performance of the classifiers using the permutation test and the area under the receiver operating characteristic curve (AUC). We analyzed the rsEEG from 50 ADHD children and 58 age-matched controls. The results show that though spectral features can be used to build a convincing model, the prediction accuracy of the model was unfortunately unstable. Bicoherence features had significant between-group differences, but classifier performance was sensitive to brain region used. rsEEG complexity of ADHD children was significantly lower than controls and may be a suitable biomarker candidate. Through a machine learning approach, 14 features from various brain regions using different methods were selected; the classifier based on these features had an AUC of 0.9158 and an accuracy of 84.59%. These findings strongly suggest that the combination of rsEEG characteristics obtained by various methods may be a tool for identifying ADHD.
Purpose
Attention-Deficit Hyperactivity Disorder (ADHD) is a neuro-developmental disorder that is characterized by hyperactivity, inattention and abrupt behaviors. This study proposes an approach for distinguishing ADHD children from normal children using their EEG signals when performing a cognitive task.
Methods
In this study, 30 children with ADHD and 30 age-matched healthy children without neurological disorders underwent electroencephalography (EEG) when performing a task to stimulate their attention. Fractal dimension (FD), approximate entropy and lyapunov exponent were extracted from EEG signals as non-linear features. In order to improve the classification results, double input symmetrical relevance (DISR) and minimum Redundancy Maximum Relevance (mRMR) methods were used to select the best features as inputs to multi-layer perceptron (MLP) neural network.
Results
As expected, children with ADHD had more delays and were less accurate in doing the cognitive task. Also, the extracted non-linear features revealed that non-linear indices were greater in different regions of the brain of ADHD children compared to healthy children. This could indicate a more chaotic behavior of ADHD children while performing a cognitive task. Finally, the accuracy of 92.28% and 93.65% were achieved using mRMR method and DISR method using MLP, respectively.
Conclusions
The results of this study demonstrate the ability of the non-linear features to distinguish ADHD children from healthy children.
Objective:
To investigate the performance of univariate and multivariate EEG measurements in diagnosing ADHD subjects in a broad age range.
Methods:
EEG from eight cortical regions were recorded at rest during eyes open and eyes closed in 22 male ADHD subjects of combined type and 21 healthy male controls (age range 4-15 years). Univariate and interdependence measurements calculated from the frequency domain and from the reconstructed state spaces of EEG signals were computed, and their performance in discriminating ADHD from healthy subjects was analyzed.
Results:
Significant between-group differences in univariate measures were age-dependent. However, certain interdependence inter-hemispheric measures during eyes closed showed significant, age-independent between-groups differences. Among them, coherence in the beta band between inter-occipital regions and between left/occipital-right/central regions provided an overall accuracy classification rate of 74.4%. Even greater accuracy (86.7%) was obtained by an interdependence index of generalized synchronization between left/occipital-right/central regions and left/central-right/temporal regions.
Conclusions:
EEG beta coherence and especially the degree of generalized synchronization between a few inter-hemispheric regions during resting state with eyes closed allow a high accuracy classification rate of ADHD subjects.
Significance:
Changes in inter-hemispheric EEG functional brain connectivity at rest are useful for ADHD diagnosis in a broad age range.