Integration of DWT, FFT, and Spatial domain for the identification of epileptic seizure utilizing electroencephalogram signal

Rabel Guharoy; Nanda Dulal Jana; Suparna Biswas; Aveek Chattopadhyaya

doi:10.2478/ijssis-2025-0041

.blurhash-client-img { display: none !important; }

Integration of DWT, FFT, and Spatial domain for the identification of epileptic seizure utilizing electroencephalogram signal

International Journal on Smart Sensing and Intelligent Systems

Volume 18 (2025): Issue 1 (January 2025)

By: Rabel Guharoy, Nanda Dulal Jana, Suparna Biswas and Aveek Chattopadhyaya

Open Access

|Aug 2025

Figures & Tables

Proposed model of epilepsy detection process. DWT, discrete wavelet transform; EEG, electroencephalogram; FFT, fast Fourier transform; LDA, linear discriminant analysis; PCA, principal component analysis.

DWT decomposition up to level two (2). DWT, discrete wavelet transform; EEG, electroencephalogram.

Samples of EEG signals of 5 different classes. EEG, electroencephalogram.

Samples of A1 after applying DWT of 5 different classes. DWT, discrete wavelet transform.

Samples of EEG signal after application of FFT for 5 different classes. EEG, electroencephalogram; FFT, fast Fourier transform.

Bar chart of total number of samples non-seizure and seizure.

Table 1:

Performance measure of case 3 (combined features)

Algorithm	Accuracy	Precision	Recall	F-1 Score
LR	0.98	0.96	0.96	0.96
K-NN	0.95	0.96	0.89	0.92
SVM	0.95	0.96	0.89	0.92

[i] K-NN, K-nearest neighbors; LR, logistic regression; SVM, support vector machine.

Table 2:

Performance measure of case 1 (combined features + PCA)

Algorithm	Accuracy	Precision	Recall	F-1 Score
LR	0.99	0.99	0.97	0.98
K-NN	0.98	0.97	0.96	0.97
SVM	0.98	0.97	0.96	0.97

[i] K-NN, K-nearest neighbors; LR, logistic regression; PCA, principal component analysis; SVM, support vector machine.

Table 3:

Performance measure of case 2 (combined features + LDA)

Algorithm	Accuracy	Precision	Recall	F-1 Score
LR	0.97	0.96	0.92	0.94
K-NN	0.96	0.96	0.90	0.92
SVM	0.96	0.96	0.90	0.92

[i] K-NN, K-nearest neighbors; LDA, linear discriminant analysis; LR, logistic regression; SVM, support vector machine.

Table 4:

Comparison with other techniques

References	Classifier	Accuracy (%)
[41]	Decision tree	99
[42]	Multilayer perceptron neural network	99
[43]	ANN	98.30
[44]	SVM	97.98
[45]	PCA with RF	92.69
[45]	PCA with ANN	97.55
Proposed model	LR	99

[i] LR, logistic regression; PCA, principal component analysis; RF, random forest; SVM, support vector machine.

References

Authors

Metrics

Articles in this issue

DOI: https://doi.org/10.2478/ijssis-2025-0041 | Journal eISSN: 1178-5608

Journal RSS Feed

Language: English

Submitted on: Mar 6, 2025

Published on: Aug 8, 2025

Published by: International Journal on Smart Sensing and Intelligent Systems

In partnership with: Paradigm Publishing Services

Publication frequency: Volume open

Keywords:

EEG,

support vector machine,

logistic regression,

K-nearest neighbors,

principal component analysis,

linear discriminant analysis

Related subjects:

Engineering,

Introductions and overviews,

Engineering, other

© 2025 Rabel Guharoy, Nanda Dulal Jana, Suparna Biswas, Aveek Chattopadhyaya, published by International Journal on Smart Sensing and Intelligent Systems
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Volume 18 (2025): Issue 1 (January 2025)

Integration of DWT, FFT, and Spatial domain for the identification of epileptic seizure utilizing electroencephalogram signal

Figures & Tables

Figure 1:

Figure 2:

Figure 3:

Figure 4:

Figure 5:

Figure 6:

Table 1:

Table 2:

Table 3:

Table 4:

Paradigm

My account