A new feature for the classification of non-stationary signals based on the direction of signal energy in the time–frequency domain
Introduction
Epilepsy is one of the most common disorders of the human nervous system and it is characterised by recurring seizures. Epileptic seizures are defined as temporary and brief disturbances in the routine electrical activity of the human brain. When too many nerve cells in the brain fire very quickly, they cause an electrical storm [36]. Because of their non-invasive nature, EEG recordings are frequently used to monitor the brain's activities. Consequently, brain abnormalities, such as seizures, can be diagnosed on the basis of deviations of the EEG recording from the normal pattern. However, the manual detection of seizures in EEG recordings requires a neurologist to be available at all times of the day and night. Therefore, there is considerable interest in the automatic detection of seizures in EEG recordings using a computer.
Machine learning algorithms are normally used in the automatic detection of seizures, which can be classified into the following categories:
- •
Model-based approaches: These approaches are based on the assumption that the signal is generated by a mathematical model. The most commonly used signal models are: the auto-regressive moving average model, the moving average model, the auto-regressive model, and the linear predictive model. Linear predictive models have been successfully employed for the detection of seizures [5,15,18].
- •
Feature-extraction approaches: Feature extraction is an operation that transforms an EEG recording into a feature vector, which represents the pattern in a reduced dimensional space. These approaches involve two stages of processing: the first stage extracts the set of features from the signal and the second stage uses the extracted features to train a classifier. The features can be extracted from the time-domain representation [30], the frequency-domain representation [16], or from joint time-frequency (TF) representations [8,19,34,37].
The extracted features are then employed by classifiers to detect the presence of a seizure in a recording. The most frequently used classifiers are: support vector machines [17], least squares support vector machine, artificial neural networks [22], genetic algorithms [19], random forest classifiers [1], and extreme learning machines [38].
EEG signals have non-stationary characteristics. This means that these signals are difficult to analyse in the time domain alone or in the frequency domain alone because their spectral content changes with time. Time–Frequency distributions (TFDs) are frequently employed for the analysis of non-stationary signals. Recently, there has been considerable interest in using TFDs for the automatic detection of changes or abnormalities in non-stationary signals, such as the detection of seizures in an EEG [21]. Although TFDs are rich in information, analysing a non-stationary signal in the TF domain increases the dimension of the signal and this complicates the signal classification problem due to the curse of dimensionality. The dimension of a TFD can be reduced by methods such as principal component analysis and singular value decomposition [9,25]. Another approach is to extract a small set of features from the TFD, which represents the same information required for discriminating between different classes of data. TF features can be broadly classified into the following categories:
- 1
Instantaneous-frequency features: Amplitude-modulation frequency-modulation (AM-FM) is a widely used model for the analysis of non-stationary signals. AM-FM signals are completely described by parameters such as the instantaneous frequency, instantaneous amplitude, and the total number of components. Therefore, instantaneous frequency (IF) based features are frequently used in the detection of seizures [20,24,31,35].
The IF can be estimated from quadratic TFDs (QTFDs) by applying sophisticated image processing algorithms [7,32] or by using signal decomposition methods, such as empirical mode decomposition or matching pursuit [27]. However, the main limitation of this approach is that estimating the IF of multi-component signals is a challenging task, especially when the IFs of the signals cross each other. Therefore, these methods are only applicable to a signal whose IFs do not overlap.
- 2
Image-based features: A TFD is a 2D representation, which is just like an image. Consequently, image related features, such as Harlic features and local binary patterns, can be extracted from a TFD [6,23,33]. These methods have shown good performance in the detection of seizure activity and in the classification of the stages of sleep [2,12].
- 3
TF features obtained by extending time-only or frequency only features: These approaches derive TF features by extending the conventional time domain and frequency domain features to the joint TF domain. Previous studies have shown that these features lead to better performance when compared to original time-only or frequency-only features [9].
In this study, a new approach to the extraction of features from a TF representation, which is based on the direction of signal energy in the TF domain, is developed and its efficacy is demonstrated using a newborn EEG database. The proposed approach first uses differentiation, which whitens the EEG background and enhances the spike-like characteristics of a seizure. A TF analysis of the EEG signal (after filtering) reveals that in the case of an EEG seizure, most of the signal energy is distributed along the frequency axis due to the spike-like signatures or along the time axis due to the tone-like characteristics. Meanwhile, in the case of the background, the signal energy is uniformly distributed along all directions [20,35]. Hence, the energy concentration of the signal distributed along the time axis or frequency axis, when normalised by the total signal energy, is considered as a feature for signal classification. Our experimental results indicate that the proposed feature outperforms in terms of the AUC, time-only, frequency-only, and commonly used TF features. It is also demonstrated that a classification system based on the proposed features achieves a total accuracy of 98.25%, which is 0.5% greater than the accuracy obtained by a combination of TF signal and image related algorithms.
Section snippets
Methodology
The complete mechanism of the proposed automatic seizure detection scheme is illustrated in Fig. 1. The important steps will be described in the following subsections.
Performance comparison of the proposed features with commonly used time-domain, frequency-domain, and time-frequency features
The proposed features are tested using the newborn EEG database, which was uploaded as a supplementary material in Ref. [10]. The database consists of 200 epochs of seizure of 200 epochs of background. Each epoch is 8 s with 256 samples acquired using a sampling frequency of 32 Hz. The area under the receiver operating characteristic curve is used as the performance criterion; namely, the probability that for a randomly chosen example, the given feature has a higher value for a seizure signal
Conclusion
A new approach to detect seizure activity in EEG signals has been proposed. This approach exploits the fact that in the presence of a seizure, the energy is concentrated along the time axes and frequency axes. Meanwhile, in the absence of a seizure, the signal energy is uniformly distributed along all directions in the TF domain. To classify non-stationary signals, such as EEG signals, based on the direction of signal energy in the TF domain, we have used the ratio of signal energy concentrated
References (38)
- et al.
Multi-category EEG signal classification developing time–frequency texture features based Fisher Vector encoding method
Neurocomputing
(2016) - et al.
Epileptic eeg detection using the linear prediction error energy
Expert Syst. Appl.
(2010) - et al.
Principles of time-frequency feature extraction for change detection in non-stationary signals: applications to newborn eeg abnormality detection
Pattern Recogn.
(2015) - et al.
Time-frequency features for pattern recognition using high-resolution tfds: a tutorial review
Digit. Signal Process.
(2015) - et al.
A comparison of computer-based methods for the determination of onset of muscle contraction using electromyography
Electroencephalogr. Clin. Neurophysiol. Electromyogr. Mot. Contr.
(1996) - et al.
An efficient, robust and fast method for the offline detection of epileptic seizures in long-term scalp eeg recordings
Clin. Neurophysiol.
(2007) - et al.
Automated analysis of brain activity for seizure detection in zebrafish models of epilepsy
J. Neurosci. Meth.
(2017) - et al.
Classification of ictal and seizure-free eeg signals using fractional linear prediction
Biomed. Signal Process Contr.
(2014) - et al.
Stockwell transform for epileptic seizure detection from eeg signals
Biomed. Signal Process Contr.
(2017) - et al.
Classification of eeg signals for epileptic seizures using hybrid artificial neural networks based wavelet transforms and fuzzy relations
Expert Syst. Appl.
(2017)
Classification of seizure and seizure-free eeg signals using local binary patterns
Biomed. Signal Process Contr.
Feature extraction and recognition of ictal eeg using emd and svm
Comput. Biol. Med.
Differential operator in seizure detection
Comput. Biol. Med.
Seizure detection algorithm for neonates based on wave-sequence analysis
Clin. Neurophysiol.
Analysis of normal and epileptic seizure eeg signals using empirical mode decomposition
Comput. Meth. Progr. Biomed.
IF estimation for multicomponent signals using image processing techniques in the time-frequency domain
Signal Process.
A new approach to characterize epileptic seizures using analytic time-frequency flexible wavelet transform and fractal dimension
Pattern Recogn. Lett.
Epileptic seizure detection based on imbalanced classification and wavelet packet transform
Seizure-European Journal of Epilepsy
Seizure detection from eeg signals using multivariate empirical mode decomposition
Comput. Biol. Med.
Cited by (30)
Local Rényi entropy-based Gini index for measuring and optimizing sparse time-frequency distributions
2024, Digital Signal Processing: A Review JournalGNMF-based quadratic feature extraction in SSTFT domain for epileptic EEG detection
2023, Biomedical Signal Processing and ControlCitation Excerpt :Due to the diversity of epileptic discharge and individual differences in EEG signals, how to find the features that can describe the EEG pattern in a better manner has ever been a challenging topic. In attempt to this burning question, numerous feature extraction algorithms including time domain [5], frequencydomain [6], time–frequency domain [7] and nonlinear dynamics analysis [8] have been developed. Some representative literature are reviewed below.
Time–frequency signal processing: Today and future
2021, Digital Signal Processing: A Review JournalCitation Excerpt :Then based to the problem definition and given classes, a set of features are extracted from the TFD. Some of the features used in many applications are the higher-order joint TF moments, and TF features which are the extended versions of time-, or frequency-domain features [79–81], and image-based features such as gray-level co-occurrence matrix, and local binary patterns [75,81–83]. Numerous computer-aided diagnosis systems based on both machine learning (ML) and deep learning (DL) have been proposed by researchers for many years to classify, detect and predict seizure activity [17,84,85].
Boosting-LDA algriothm with multi-domain feature fusion for motor imagery EEG decoding
2021, Biomedical Signal Processing and ControlCitation Excerpt :Early EEG analysis is to directly extract significant waveform features, such as mean value and peak value, etc. [10,11]. The subsequent frequency domain analysis method directly observes the EEG rhythm distribution from the perspective of frequency domain to extract features [12,13]. With the progress of digital signal processing technology, time–frequency characteristic analysis has been applied to the feature extraction of EEG signals, and a good effect has been achieved by replacing the single-scale analysis method with the dual resolution of time and frequency [14,15].
An instantaneous frequency and group delay based feature for classifying EEG signals
2021, Biomedical Signal Processing and ControlCitation Excerpt :We employ the area under the receiver operating curve (AUC) as a criterion. Table 2 illustrates that the best performance is obtained by the energy concentration of the directionally filtered time-frequency distribution (ECDTFD) [29] followed by the proposed feature. Note that the computation of ECDTFD is very expensive as it involves the computation of the adaptive directional time-frequency distribution (ADTFD) with the computational complexity of O(N2 log N + RP2N2), where P is the size of the smoothing filter, N is the number of samples in a signal and R is the number of quantization levels [47].