Classification of Normal and Epileptic EEG Signal Using Time &amp;amp;amp; Frequency Domain Features through Artificial Neural Network

Anusha, K.; Mathews, Mathew T.; Puthankattil, Subha D.

doi:10.1109/icacc.2012.21

Cited by 16 publications

(4 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some methods have been proposed to detect epileptic seizures and these methods are divided into traditional and end-to-end methods according to whether manual feature extraction is performed or not. In traditional methods, EEG data is first processed with feature extraction and then classified by using some classifiers [13][14][15][16]. In the end-to-end method [17,18], feature extraction and classification are fused into one step by using the machine learning model.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Epileptic seizure detection by using interpretable machine learning models

et al. 2023

View full text Add to dashboard Cite

Objective. Accurate detection of epileptic seizures using electroencephalogram (EEG) data is essential for epilepsy diagnosis, but the visual diagnostic process for clinical experts is a time-consuming task. To improve efficiency, some seizure detection methods have been proposed. Regardless of traditional or machine learning methods, the results identify only seizures and non-seizures. Our goal is not only to detect seizures but also to explain the basis for detection and provide reference information to clinical experts. Approach. In this study, we follow the visual diagnosis mechanism used by clinical experts that directly processes plotted EEG image data and apply some commonly used models of LeNet, VGG, deep residual network (ResNet), and vision transformer (ViT) to the EEG image classification task. Before using these models, we propose a data augmentation method using random channel ordering (RCO), which adjusts the channel order to generate new images. The Gradient- weighted class activation mapping (Grad-CAM) and attention layer methods are used to interpret the models. Main results. The RCO method can balance the dataset in seizure and non-seizure classes. The models achieved good performance in the seizure detection task. Moreover, the Grad-CAM and attention layer methods explained the detection basis of the model very well and calculate a value that measures the seizure degree. Significance. Processing EEG data in the form of images can flexibility to use a variety of machine learning models. The imbalance problem that exists widely in clinical practice is well solved by the RCO method. Since the method follows the visual diagnosis mechanism of clinical experts, the model interpretation results can be presented to clinical experts intuitively, and the quantitative information provided by the model is also a good diagnostic reference.

show abstract

Section: Discussionmentioning

confidence: 99%

“…However, the frequency features change dramatically in seizure EEG data. Some methods utilize the time-frequency feature, and the shorttime Fourier transform and wavelet transform are commonly used [13][14][15]. In the next step, the extracted features are sent to common classifiers for the classification task to achieve seizure detection.…”

Section: Introductionmentioning

confidence: 99%

Epileptic seizure detection by using interpretable machine learning models

et al. 2023

View full text Add to dashboard Cite

show abstract

“…As a consequence of this, heart rate calculations are highly erroneous [6,7]. As a consequence of pacemaker artefacts, automatic heart rate estimate may be highly erroneous [8,9,10]. The presence of noise in the ECG signal results in erroneous heart rate and heart rate variability (HRV) measurements.…”

Section: Introductionmentioning

confidence: 99%

Manoeuvre a Cross-Spectral Algorithm and Machine Learning Approach to Categorize Seizures

Ojo

Allabun

Shukla

et al. 2023

Preprint

View full text Add to dashboard Cite

EEG signals are recordings of electrical activity in the brain, and they can be used to detect epileptic seizures. Cross-spectral analysis involves analyzing the spectral coherence between different frequency bands in EEG signals. Cross-spectral seizure detection is a technique used to detect seizures in electroencephalography (EEG) signals by analyzing the frequency content of the signal in different frequency bands. Spectral coherence is a measure of how correlated two signals are in the frequency domain, and it can be used to identify patterns in EEG signals that are characteristic of seizures. Even if the ECG, which provides a direct measure of heart rate, were polluted by noise or missing altogether, the heart rate computed from such signals would be erroneous. This necessitates the use of an accurate heart rate estimate, which is especially important when the ECG is noisy or missing. To put it another way, fusion combines cardiovascular data with no cardiovascular (NC) signals, which are not connected to cardiac activity but include signs of heartbeats. According to the results of our evaluation of standard datasets, they determined that the SSF-TKE approach is particularly successful at identifying R-peak artefacts in non-cardiovascular signals that ECG anomalies have contaminated. When tested on standard datasets, the beat SQI-based voting fusion technique demonstrated a high degree of accuracy in predicting heart rate from a fusion of multimodal information. Compared to the single signal technique, the fusion methodology out per-forms it when determining heart rate precision. This strategy was evaluated using ECG and ABP signals from a synthetic noise dataset, which was created by adding various forms of calibrated noise to clean signals and then testing the outcomes of the technique on those signals. As a result of our paper, we noticed that merging cardiovascular and no cardiovascular inputs increased the accuracy of physiological parameter assessment.

show abstract

“…Anusha et al (2012) andTzallas et al (2017) use a similar method in the pre-processing stage for experiments to classify normal and epileptic EEG signals using Artificial Neural Network (ANN) and Brain-Computer Interface, respectively.…”

mentioning

confidence: 99%

Automated classification of eight different Electroencephalogram (EEG) bands using hybrid of Fast Fourier Transform (FFT) with machine learning methods

Noor¹,

Malim

Rostam

et al. 2022

Neurosci Res Notes

View full text Add to dashboard Cite

Analysing and processing the EEG dataset is crucial. Countless actions have been taken to ensure that the researcher in brain studies always achieves informative data and produces notable findings. There are several standard procedures to produce an informative result in analysing the EEG data. However, the techniques used in each standard procedure might be different for the researcher or data analyst because they have their preferences to suit the purpose of their experiments to adapt with the dataset collected. Not only the current manual method is time-consuming, but the main challenges are that researchers need to analyse only a small portion of the brain signals that are the most relevant to be observed through the analysis of several bands such as Very low, Delta, Theta, Alpha-1, Alpha-2, Beta-1, Beta-2, and Gamma. Therefore, one of the best alternatives is to automate the process of classifying the eight bands and extract the most relevant features. Hence, this paper proposed an automated classification method and feature extraction method through hybridising Fast Fourier Transform (FFT) with three different machine learning methods (KNN, SVM, and ANN) that can improve the efficiency of EEG analysis. Based on the result, the FFT + SVM method gives a 100% accuracy and successfully classified the bands into different of eight EEG bands accurately.

show abstract

Classification of Normal and Epileptic EEG Signal Using Time &amp; Frequency Domain Features through Artificial Neural Network

Cited by 16 publications

References 1 publication

Epileptic seizure detection by using interpretable machine learning models

Epileptic seizure detection by using interpretable machine learning models

Manoeuvre a Cross-Spectral Algorithm and Machine Learning Approach to Categorize Seizures

Automated classification of eight different Electroencephalogram (EEG) bands using hybrid of Fast Fourier Transform (FFT) with machine learning methods

Contact Info

Product

Resources

About