**2.3 Dynamic mode decomposition**

In fluid flow analysis studies, generally, computationally expensive Global stability analysis method where classical approaches are used, is performed. Proper Orthogonal Decomposition (POD) method based on snapshots of flow and achieving the most active modes is used in these methods. DMD approach based on matrix decomposition has been proffered as a solution to computationally cost of these previous approaches. Systems are analyzed in space using the DMD method in which temporal orthogonality is used. However, using the POD method utilizing spatial orthogonality, systems could be analyzed in time [30]. The behavior of nonlinear and dynamic systems such as biological signals cannot be completely revealed by classical time-frequency analysis methods. By evaluating the measurements collected over a certain period of time with the DMD method, both the system can be expressed with a function, and information about the future behavior of the system can be predicted. The basic idea of the DMD method is to obtain the dynamic modes that best represent the system by achieving the eigenvalues and eigenvectors of the system that linearized with the Least-Squares Approximation (LSA) method [31, 34].

In literature, previously, *K* � *T*� sized multi-channel EEG signals are evaluated using the DMD approach. Here, T is the sample size of a single EEG channel, and N is the number of channels. Using this data matrix, *K* � *L*� sized *X* data matrices in which L denotes the time samples named "snapshot" is obtained, and the DMD algorithm is applied to this obtained data matrices [31]. In our study, both the multi-channel DMD approach used in the literature is performed and the singlechannel DMD approach is proposed, unlike the literature, and *K* � *L*� sized *X* data matrices are constructed using this two different approaches.

In the **single-channel DMD approach (SC-DMD)**, the single-channel EEG signals with *T*� samples long are divided into non-overlapping, *L* samples long EEG segments. The ð Þ *K* � *L* EEG data matrices are constructed using *K* of these obtained segments. For our epileptic seizure classification experiment, *L* ¼ 140 and *K* ¼ 5 are chosen.

Additionally, in the **multi-channel DMD approach (MC-DMD)**, ð Þ *K* � *L* EEG data matrices with no overlap are generated using *L* ¼ 140 samples of *K* ¼ 5 different EEG channels. In our experiment, these data matrices are obtained using the *K* ¼ 5�EEG channel in the left hemisphere (Fp1-F7, F7-T1, T1-T3, T3-T5, Fp1- F3) and the *K* ¼ 5�EEG channel in the right hemisphere (Fp2-F8, F8-T2, T2-T4, T4-T6, Fp2-F4). Also 10 ð Þ � 120 EEG data matrices are constructed using the *K* ¼ 10�EEG channel with *L* ¼ 120 sample long in both hemispheres.

In order to achieve a sufficient number of modes to demonstrate the dynamics of neurological activity efficiently, the number of ð Þ *K* measurements must be at least twice the number of *L* time points named snapshots [16]. Therefore, the data augmentation process is applied to the data matrix *X* based on the Hankel matrix creation principle as detailed in [34] and the *N* � *M*,ð Þ *N* ≥2*M* dimensional augmented data matrix *Xa* is obtained.

*Epilepsy - Update on Classification, Etiologies, Instrumental Diagnosis and Treatment*

$$X\_a = \begin{bmatrix} \vdots & \vdots & \dots & \vdots \\ \infty\_1 & \infty\_2 & \dots & \infty\_{M-1} \\ \vdots & \vdots & \dots & \vdots \end{bmatrix} \qquad X'\_a = \begin{bmatrix} \vdots & \vdots & \dots & \vdots \\ \infty\_2 & \infty\_3 & \dots & \infty\_M \\ \vdots & \vdots & \dots & \vdots \end{bmatrix} \tag{4}$$

$$X'\_a = A X\_a \tag{5}$$

Transition matrix A that denoted in Eq. (5) should be obtained to achieve relation based on the high-dimensional linear regression between *Xa* matrix and its time-shifted version *X*<sup>0</sup> *<sup>a</sup>* matrix (given in Eq. (4)). This transition matrix can be calculated using the pseudo-inverse of the *Xa* matrix *A* ¼ *X*<sup>0</sup> *aX*<sup>þ</sup> *a* � �, but for higherdimensional data such as biosignal, this can cause computational complexity. Using the DMD algorithm;

Singular value decomposition (SVD) of augmented data matrix *Xa* <sup>¼</sup> *<sup>U</sup>*Σ*<sup>V</sup>* <sup>∗</sup> is calculated, and formulation of transition matrix rewrite again using the Left singular vectors *U*, the inverse of the singular values Σ�<sup>1</sup> , and the Right singular vectors *V A* ¼ *X*<sup>0</sup> *aX*<sup>þ</sup> *<sup>a</sup>* ¼ *X*<sup>0</sup> *aV*Σ�<sup>1</sup> *U* <sup>∗</sup> . The low-rank approximation value *A*~ of the transition matrix *A* can be obtained using Eq. (6)

$$
\tilde{A} = U^\* X\_a' V \Sigma^{-1} \tag{6}
$$

The Eigen decomposition of *<sup>A</sup>*<sup>~</sup> matrix is calculated (*AW*<sup>~</sup> <sup>¼</sup> *<sup>W</sup>*Ω) and the matrix of eigenvectors *W*, the diagonal matrix Ω of eigenvalues are achieved. Finally, DMD modes of augmented data matrix *Xa* are calculated using Eq. (7) where each column of *ϕ* includes the DMD mode *ϕ<sup>m</sup>* related to eigenvalues *λ<sup>m</sup>* [31, 34].

$$
\phi = X\_a' V \Sigma^{-1} W \tag{7}
$$

In our DMD based epileptic seizure classification experiment, using the DMD spectrum, various features based on DMD subband powers and Higher-order DMD spectral moments (DMD-HOS) are calculated and classification performances of approaches are compared.

The real part, of DMD modes associated with the complex eigenvalues *λm*, indicates the decay frequency of the dynamic modes, while the imaginary part of these modes shows the oscillation frequencies of the dynamic modes. To obtain the DMD spectrum of pre-seizure and seizure EEG segments, oscillation frequencies, and powers, of the dynamic modes, should be calculated. The oscillation frequencies *f <sup>m</sup>* (*Hz*) are calculated using Δ*t* ¼ 0*:*01*s* time difference between sequential snapshots, and the complex eigenvalues *<sup>λ</sup><sup>m</sup>* of DMD modes; *<sup>f</sup> <sup>m</sup>* <sup>¼</sup> <sup>∣</sup>*imag <sup>ω</sup><sup>m</sup>* 2*π* � �∣, *<sup>ω</sup><sup>m</sup>* <sup>¼</sup> *log* ð Þ *<sup>λ</sup><sup>m</sup>* <sup>Δ</sup>*<sup>t</sup>* (the imaginary part of a complex number is calculated using *imag*ð Þ*:* operation). The frequency set *FDMD* ¼ *f <sup>m</sup>* � � is obtained by aligning the oscillation frequencies containing different mode frequencies. Additionally, power *Pm* ¼ k k *ϕ<sup>m</sup>* <sup>2</sup> of these modes are calculated using the Euclidian norm [34]. The total DMD mode power *f <sup>m</sup>* � �∈*FDMD* (given in Eq. (8)) for the *f <sup>m</sup>* frequency is calculated by summing the power value of *Lk* DMD modes at the *f <sup>m</sup>* frequency. This process is repeated for all frequencies in the *FDMD* set and a single DMD power corresponding to each frequency is calculated. In order to obtain the **DMD spectrum**, the obtained DMD power set *PDMD*, ∀ *PDMD f <sup>m</sup>* � � � � ∈*PDMD* is plotted according to the oscillation frequency set *FDMD*.

$$P\_{\text{DMD}}\left(\boldsymbol{f}\_{m}\right) = \sum\_{i=1}^{L\_{k}} P\_{m}^{i}\left(\boldsymbol{f}\_{m}\right) \qquad \forall \left\{\boldsymbol{f}\_{m}\right\} \in F\_{\text{DMD}}.\tag{8}$$

*Epileptic EEG Classification by Using Advanced Signal Decomposition Methods DOI: http://dx.doi.org/10.5772/intechopen.93810*

To reveal the advantages of the DMD approach, the traditional Power Spectral Density is estimated using the Welch method [5, 35] where the Hamming window and 50% overlapping are chosen, for each seizure, and pre-seizure EEG segments (140 samples long = 1.4 sec). Examples of the proposed Single-Channel EEG based DMD spectra and traditional Welch PSD estimates for pre-seizure and seizure epileptic EEG data are demonstrated in **Figure 2**. The similarity between the average PSD values of the 5 EEG segments (shown with bold black lines in **Figure 2(c)** and **(f)**) whose PSDs are calculated separately by the Welch method and the DMD spectrum, given in **Figure 2(b)** and **(e)**, is remarkable.

In DMD based epileptic seizure detection approach, sub-band powers based and DMD-HOS moments based features are introduced using the DMD spectrum. In computer-aided epileptic seizure detection and prediction studies, EEG subband powers of different frequency bands like delta (0–4 Hz), theta (4–8 Hz), alpha (8–12 Hz), beta (12–30 Hz), and gamma (30–60 Hz), and DMD-HOS moments are calculated using conventional Power Spectral Density [17, 40]. Using the estimated DMD spectrum, similar to the classical PSD approach, Delta (*Pδ*), Theta (*Pθ*), Alpha (*Pα*), Beta (*Pβ*), and Gamma (*Pγ*)) **subband powers** are calculated as

$$P\_{sb} = \sum\_{f\_m \in f\_{sb}} P\_{\text{DMD}} \left( f\_m \right), \quad sb = \{ \delta, \theta, a, \beta, \gamma \} \qquad P\_T = \sum\_{f\_m} P\_{\text{DMD}} \left( f\_m \right) \tag{9}$$

We propose another set of features called **DMD-HOS moments** *Mj DMD*, *j* ¼ 1, 2, 3, ⋯ defined by

$$M\_{\rm DMD}^j = \sum\_{f\_m \in F\_{\rm DMD}} (f\_m)^j P\_{\rm DMD}(f\_m), \qquad j = 1, 2, 3, \dots \tag{10}$$

In Eqs. (9) and (10), *fsb* is a subset of oscillation frequencies in *FDMD* ¼ *f <sup>m</sup>* � � of extracted DMD modes corresponding to sub-band frequencies *f <sup>δ</sup>*, *f <sup>θ</sup>*, *f <sup>α</sup>*, *f <sup>β</sup>*, *f <sup>γ</sup>* n o

#### **Figure 2.**

*First 5 DMD modes of 5 pre-seizure EEG segments (a) and its DMD spectrum (b) obtained using Single Channel EEG based dynamic mode decomposition, PSD of these 5 pre-seizure EEG segments together (c); first 5 DMD modes of 5 seizure EEG segments (d) and its DMD spectrum (e) obtained using Single Channel EEG based dynamic mode decomposition, PSD of these 5 seizure EEG segments together (f). Bold black lines denote the average of 5 PSD in (c) and (f).*

of EEG, *PT* denotes the total power of DMD spectrum, and *M<sup>j</sup> DMD* indicates the *j th* order DMD spectral moment. In our computations, we extract 6 DMD subband power-based features, and 3 DMD-HOS moments features for each seizure and pre-seizure EEG segment.

#### **2.4 Synchrosqueezing transform**

Synchrosqueezing Transform is a member of TF reassignment methods (RM) family which developed to improve the localization properties of TFRs. In RM methods, using the reassignment process, TF coefficients *X t*ð Þ ,*ω* that computed utilizing classical TF analysis method, are reassigned into the instantaneous frequency (IF) trajectory close to the ideal TFR which have high frequency and time resolution ð Þ *t*,*ω* ↦ ð Þ *τ*0ð Þ *t*, *ω* , *ω*0ð Þ *t*, *ω* . On the other hand, using the squeezing process, this TF coefficients *X t*ð Þ ,*ω* are squeezed into the IF trajectory close to the ideal TFR which have high-resolution in only frequency ð Þ *t*, *ω* ↦ ð Þ *t*, *ω*0ð Þ *t*, *ω* . Although lower TF resolution is achieved using the SST method, signal reconstruction may be performed [29, 36].

SST method based on STFT or CWT can be performed to obtain high-resolution TFRs of signals. Hence, the TF coefficients of the studied signals are obtained by STFT or CWT, and by using these coefficients with the SST approach, high-resolution TFR is obtained.

In the STFT method, the signal is divided into short-time, and usually overlapping segments and the Fourier transforms of these short-term segments are calculated. In our computations, STFT of 1-second EEG segment *x t*ð Þ, are calculated as, *X t*ð Þ¼ ,*<sup>ω</sup>* <sup>Ð</sup> <sup>∞</sup> �∞*x*ð Þ*<sup>τ</sup> <sup>w</sup>*ð Þ *<sup>τ</sup>* � *<sup>t</sup> <sup>e</sup>*�*jωτd<sup>τ</sup>* where *w t*ð Þ denotes the used window function. Using the Fourier transforms of analyzed segment *X*ð Þ *ω* and used window function *W*ð Þ *ω* , STFT may be rewritten again as given in Eq. (11).

$$X(t,\omega) = \frac{1}{2\pi} \int\_{-\infty}^{\infty} X(\xi)W(\omega - \xi)e^{j\xi t}d\xi. \tag{11}$$

In the SST approach, computing the derivative of STFT *X t*ð Þ , *ω* according to time, the instantaneous frequency *ω*0ð Þ¼� *t*, *ω j <sup>∂</sup>tX t*ð Þ ,*<sup>ω</sup> X t*ð Þ , *<sup>ω</sup>* is obtained. By using synchrosqueezing operator Ð <sup>∞</sup> �<sup>∞</sup>*δ η*ð Þ � *<sup>ω</sup>*0ð Þ *<sup>t</sup>* � *<sup>ω</sup> <sup>d</sup><sup>ω</sup>* of SST and IF *<sup>ω</sup>*0(t, *<sup>ω</sup>*), SST *T t*ð Þ , *η* with high-resolution is obtained by collecting the STFT coefficients which have the same frequency where they should appear.

$$T(t,\eta) = \int\_{-\infty}^{\infty} X(t,\alpha)\delta(\eta - \alpha\_0(t - \alpha))d\alpha \tag{12}$$

An example TF representations of 1-sec pre-seizure and seizure EEG segments achieved utilizing SST and STFT approaches are shown in **Figure 3**. We observe in **Figures 3(b)**, **(c)**, **(e)** and **(f)** that the SST approach is able to represent pre-seizure and seizure EEG segments better in the TF plane than the STFT method. Although the window size, which is the most important parameter of STFT [19], is chosen to give the best time and frequency resolution, the SST approach provided better TF resolution.

In our SST based epileptic seizure detection study, high-resolution joint TF distributions of pre-seizure and seizure EEG segments are calculated. Two different feature extraction approaches are presented to achieve efficient features from the magnitude square of the SST matrix *S n*ð Þ ,*ω<sup>k</sup>* :

*Epileptic EEG Classification by Using Advanced Signal Decomposition Methods DOI: http://dx.doi.org/10.5772/intechopen.93810*

a. Log-normalized **higher-order joint TF (HOJ-TF) moments**, <*ni ω<sup>j</sup> <sup>k</sup>* > ; *i*, *j* ¼ 1, 2, … [37], <*ni ω<sup>j</sup> <sup>k</sup>* > ¼ log P*N*�<sup>1</sup> *n*¼0 P*N*�<sup>1</sup> *<sup>k</sup>*¼<sup>0</sup> *ni ωj <sup>k</sup>S n*ð Þ , *ω<sup>k</sup>* !, *i*, *j* ¼ 1, … (13)

*i*!*j*!

where *<sup>N</sup>* is the length of the EEG segments, and *<sup>ω</sup><sup>k</sup>* <sup>¼</sup> <sup>2</sup>*<sup>π</sup> <sup>N</sup> k*, *k* ¼ 0, … , *N* � 1.

b. TFR obtained by SST is used as image and **Gray Level Co-occurrence Matrix (GLCM)** texture descriptors are obtained from this TFR image.

GLCM is a prediction of the joint probability distribution of two neighboring gray-level image pixel pairs with a certain position that consists of distance (*d*) and direction (*θ*) information. The GLCM of this image can be expressed as given in Eq. (14) using image pixel pair position information (Δ ¼ ð Þ *θ*, *d* ).

$$G\_{\Delta}(i, j) \qquad \left(i, j = 0, 1, \ldots, N\_{\mathcal{g}} - 1\right) \tag{14}$$

where, *i* and *j* indicate the intensity values of two pixels, and *Ng* is the number of gray levels in the image [28, 38]. Second-order statistical features such as contrast, correlation, energy, and homogeneity [39] are calculated as features from the GLCM matrix of TF images corresponding to pre-seizure and seizure EEG segments. In order to evaluate the performance of the SST approach, same features are calculated using the magnitude square of STFT, i.e., the spectrogram j j *X t*ð Þ , *<sup>ω</sup>* <sup>2</sup> , which is a classical TF approach and is also used in the SST algorithm [19]. In our experiments, 3 HOJ-TF moments based features, and 4 GLCM based features are calculated for each pre-seizure and seizure EEG segment using both SST and STFT approaches.

**Figure 3.**

*(a) 1-sec pre-seizure EEG segment, its (b) magnitude SST, and (c) magnitude STFT; (d) 1-sec seizure EEG segment, its (e) magnitude SST, and (f) magnitude STFT.*

#### *2.4.1. Classification and performance evaluation*

In the proposed study, features extracted utilizing the three different approaches are classified using six different classifiers such as SVM, kNN, NB to distinguish seizure and pre-seizure EEG segments.

In SVM, one of the well-known supervised learning algorithms, decision boundaries, called hyper-planes', is determined to categorize data. While there are many possible hyper-planes that may be constructed, it is essential to determine the hyper-plane where the best classification performance is obtained. The optimal hyper-plane is achieved by maximizing the margin, which is the distance between different classes'support vectors. Once the optimum hyper-plane is determined, the data falling on different sides of the hyper-plane are assigned as elements of different classes. While this process used for the only linearly separable datasets, using the kernel functions SVM is performed to distinguish linearly non-separable datasets. In our proposed study, linear kernel function is performed [10, 13].

The basic and efficient machine learning method kNN is one of the most widely used supervised learning approaches. The distance between each sample *x*<sup>0</sup> to be classified in the test data and the training data is calculated for all data set which is randomly divided into tests and trains. By determining the k neighbors that have the minimum distance, the most common class among these k neighbors is assigned as the class of this sample. Although there are various distance calculation metrics such as Euclidean, Manhattan, Minkowski, and Hamming, the Euclidean distance metric, which is the most commonly used in the literature, is used in our study. In addition, the value of k is chosen as 10 for the proposed study [39, 40].

The NB classifier is one of the probabilistic classifiers in which the classification is performed according to Bayes' theorem. Membership probabilities *P M*ð Þ *<sup>i</sup>=x*<sup>0</sup> (*Mi* indicates the class, *c* denotes the number of class) to "c" classes of sample *x*<sup>0</sup> to be classified are calculated, separately. This sample is assigned as a member of the class in which the highest probability of membership among the "c" class is calculated [39, 40].

To achieve a stable classification accuracy, 10–fold cross-validation is employed in our experiment. Using Accuracy (ACC), and *F*1-score metrics, performance evaluation of proposed methods and utilized classifiers are investigated.

$$\text{ACC} = \frac{\text{TP} + \text{TN}}{\text{TP} + \text{FN} + \text{FP} + \text{TN}}, \quad F\_{\text{Score}} = 2 \ast \frac{\text{PRE} \ast \text{REC}}{\text{PRE} + \text{REC}} \tag{15}$$

where true-positive (TP) is the number of samples of *class*\_1 classified in the same class, and true-negative (TN) denotes the number of samples of *class*\_0 classified in *class*\_0. While false-positive (FP) is the number of samples not in *class*\_1 but classified in *class*\_1, false-negative (FN) indicates the number of samples in *class*\_1 but classified in *class*\_0. Recall and Precision metrics are formulated respectively as, *REC* <sup>¼</sup> *TP TP*þ*FN*, and *PRE* <sup>¼</sup> *TN FP*þ*TN* [18].

#### **3. Experiments and results**

In the following, we give the performance evaluation of seizure and pre-seizure EEG classification by using three different advanced signal representation methods presented in Section 2. The classification process is performed using SVM, kNN, and NB classifiers and compared the performance of each approach and classifier. In the **Tables 1**–**5**, highest classification performances are indicated with boldface numbers for each approach and component.


*Epileptic EEG Classification by Using Advanced Signal Decomposition Methods DOI: http://dx.doi.org/10.5772/intechopen.93810*

#### **Table 1.**

*Performance results (%) of EMD and EEMD based seizure detection approach.*


**Table 2.**

*Performance results (%) for seizure detection using the subband power based feature set of DMD based approach.*

#### **3.1 Results of EMD methods**

In EMD and EEMD based seizure detection approaches, various features in the time-domain, spectral-domain, and non-linear are calculated to separate the seizure and pre-seizure EEG segments. To compare the performances of EMD based approaches, DWT approach is implemented to the pre-seizure and seizure EEG


#### **Table 3.**

*Performance results (%) for seizure detection using the DMD-HOS moment based feature set of DMD based approach.*


#### **Table 4.**

*Performance results (%) for seizure detection using the HOJ-TF moment based feature set of SST and STFT based approaches.*


#### **Table 5.**

*Performance results (%) for seizure detection using the GLCM based feature set of SST and STFT based approaches.*

segments, and same features are calculated from the Approximation Coefficient (AC) and 3 Detail Coefficients (DC) of DWT. Additionally, without using any signal processing approach the same features are extracted from the EEG signals itself.

*Epileptic EEG Classification by Using Advanced Signal Decomposition Methods DOI: http://dx.doi.org/10.5772/intechopen.93810*

The performance evaluation results for different IMF combinations are demonstrated in **Table 1**. In all tables, we indicate the highest classification performance with boldface numbers for each case. In **Table 1**, the components column shows that the features for classifications are calculated by using the corresponding component. For example, the classification results of the features calculated using IMF1 are given in the first row, and the classification results of the features calculated from the EEG signal itself are given in the last row. NB classifier provides the highest classification successes for both EMD (96.88% ACC, 96.77% *F*1-score) and EEMD (97% ACC, 96.91% *F*1-score) approaches by using features calculated from IMF1-IMF3 (the first two IMFs decided by the IMF selection process) of the corresponding approach. While, the maximum (94.56% ACC, 94.43% *F*1-score) classification successes are achieved using the NB classifier for the DWT approach; using the kNN classifier and EEG signals itself, maximum (93.25% ACC, 93.35% *F*1 score) values are obtained.

#### **3.2 Results of DMD methods**

Performance evaluation results of SC-DMD and MC-DMD based and PSD based epileptic seizure detection approaches are summarized in **Tables 2**–**3**. For the SC-DMD and PSD approaches, the classification results of the feature set created by combining the features obtained from the Left Hemisphere (Fp1-F7, F7-T1, T1-T3, T3-T5, Fp1-F3), Right hemisphere (Fp2-F8, F8-T2, T2-T4, T4-T6, Fp2-F4), and both hemisphere (Fp1-F7, F7-T1, T1-T3, T3-T5, Fp1-F3, Fp2-F8, F8-T2, T2-T4, T4- T6, Fp2-F4) channels separately are denoted with "Left Hems", "Right Hems "and "Two Hems", while the same components show the classification results of DMD features obtained from the EEG data matrix created using the respective hemisphere channels in the MC-DMD approach.

For all three approaches, the highest classification performance for both the subband based feature set and the moment based feature set is obtained from the Left Hems. While the kNN classifier is yield to highest classification accuracy 94.1% and *F*1-score 95.5% for subband power-based feature set obtained from the Left Hems of SC-DMD approach, the maximum 92.2% ACC and 93.9% *F*1-score values are achieved with the SVM classifier using the moment-based feature set of the SC-DMD approach. On the other hand, in the MC-DMD approach, the classification performances of subband power-based (kNN: 93.9% ACC, 95.5% *F*1-score) and moment-based (SVM: 92.9% ACC, 93.4% *F*1-score) feature sets are close to each other for Left Hems and Two Hems. Additionally, using the PSD approach, a maximum of 92.2%, and 92.5% classification accuracies are achieved using the kNN and SVM classifiers for the subband power-based and moment-based feature sets of Left Hems, respectively. The results show that both SC-DMD and MC-DMD approaches are more successful than the classical PSD approach.

#### **3.3 Results of SST and STFT methods**

Performance evaluation results of the SST based approach are given in **Tables 4** and **5**. Analyzes for SST and STFT approaches are carried out separately for each channel. The classification result of the feature set created by combining the features obtained from the left hemisphere channels is given with the "Left Hems" component. Similarly, while the classification result of the feature set obtained for the right hemisphere is given with "Right Hems", the classification result of the feature set created by combining the features obtained from all channels is given with the "Two Hems" component.

Classification performance of HOJ-TF based feature set is higher than that of GLCM based feature set for each component of SST and STFT approaches. In both approaches, the classification success of both the HOJ-TF moment based feature set and the GLCM based feature set is higher in Left Hems than in Right Hems. While the highest 93.1% ACC and 94.6% *F*1-score are provided with SVM classifier by using the HOJ-TF moment-based feature set for Left Hems of SST, the maximum 92.6% ACC and 94.2% *F*1-score are obtained with the kNN classifier using the GLCM based feature set. On the other hand, in the STFT approach, 92.1% ACC and 93.8% *F*1-score values are achieved with the SVM using the HOJ-TF moment-based feature set, while the classification performance of GLCM based feature set is 90.4% ACC, and 92.4% *F*1-score for kNN classifier.

*F*1-scores obtained by the proposed methods, and by the classical approaches are calculated for comparison and given in **Figure 4**. The *F*1-scores of the proposed EMD and EEMD-based approaches, in **Figure 4a**, are higher than those of DWT and EEG-based approaches, except for the kNN classifier. In the DMD-based seizure detection approach, higher *F*1-score values are obtained in all classifiers than that of the traditional PSD approach for the subband power-based feature set, while the DMD approach provided higher *F*1-score values in the moment-based feature set, except for the NB classifier, shown in **Figure 4b**. Finally, in the SST-based epileptic seizure detection approach, higher *F*1-score values are obtained for each feature set and classifier compared to the STFT approach as shown in **Figure 4c**.

Channel-based classification performances of the proposed SC-DMD, SST, EMD, and EEMD approaches are given with a topographic maps in **Figure 5**. The topographic map is created by averaging the ACC values obtained with all classifiers for each method. It was stated by the expert neurologists that epileptic attacks in the used data set are left hemisphere-focused. It is noteworthy that the channel-based classification success of the EEG-based seizure detection approach (shown in **Figure 5a**) is very low, while is very high for the EEMD-based seizure detection approach (given in **Figure 5c**). It is also remarkable that in all proposed methods, the channels in the left hemisphere yielded successful results of seizure detection (given in **Figure 5b**–**5e**).

#### **Figure 4.**

*Changing of F*1*-score values of (a) EMD and EEMD based, (b) DMD based, and (c) SST and STFT based epileptic seizure detection approaches.*

*Epileptic EEG Classification by Using Advanced Signal Decomposition Methods DOI: http://dx.doi.org/10.5772/intechopen.93810*

**Figure 5.**

*Topographic map of channel based classification accuracies of (a) EEG based (b) EMD based (c) EEMD based (d) SC-DMD based, and (e) SST based approaches.*
