A CNN-LSTM hybrid network for automatic seizure detection in EEG signals

Shanmugam, Shalini; Selvathi, D.

doi:10.1007/s00521-023-08832-2

Cited by 21 publications

(5 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Notably, the performance of the method in Pandey et al (2023) was not assessed for the model’s generalization ability and was limited to a three-category classification problem in their literature. In the five-category classification task, the method ( Shanmugam and Dharmar, 2023 ) achieves the highest accuracy, closely followed by our model, with an epilepsy recognition accuracy of 92.50, 1.23% higher than our proposed model. However, our model’s accuracy in the AB-CD-E classification task exceeds that of Shanmugam and Dharmar (2023) by 2.15%.…”

Section: Discussionmentioning

confidence: 79%

“…In the five-category classification task, the method ( Shanmugam and Dharmar, 2023 ) achieves the highest accuracy, closely followed by our model, with an epilepsy recognition accuracy of 92.50, 1.23% higher than our proposed model. However, our model’s accuracy in the AB-CD-E classification task exceeds that of Shanmugam and Dharmar (2023) by 2.15%. Therefore, our model exhibits superior efficacy and adaptability in various epilepsy recognition and classification tasks compared to other methods.…”

Section: Discussionmentioning

confidence: 79%

“…A total of 14 recognition tasks on Bonn dataset were performed, comprising eight two-category tasks, five three-category tasks, and one five-category task. Table 5 presents a comparison with recent DL approaches from the literature, including an RNN-based method ( Tuncer and Bolat, 2022b ), and several hybrid-based methods ( Xu et al, 2020 ; Ahmad et al, 2023 ; Pandey et al, 2023 ; Qiu et al, 2023 ; Shanmugam and Dharmar, 2023 ; Wang et al, 2023 ).…”

Section: Discussionmentioning

confidence: 99%

“…Their method achieved recognition accuracies of 99.39 and 82.00% for two-category and five-category epileptic seizure detection tasks, respectively, on the Bonn university epilepsy dataset. Furthermore, several analogous studies have utilized varying numbers of convolutional layers or LSTM layers ( Yang et al, 2021 ; Pandey et al, 2023 ; Shanmugam and Dharmar, 2023 ; Wang et al, 2023 ; Yu et al, 2023 ). Similarly, Ahmad et al (2023) combined CNN and BiLSTM in a series, implementing truncated backpropagation through time to efficiently capture spatial and temporal sequence information while minimizing computational complexity.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Residual and bidirectional LSTM for epileptic seizure detection

Zhao,

Wang,

Patnaik

et al. 2024

Front. Comput. Neurosci.

View full text Add to dashboard Cite

Electroencephalogram (EEG) plays a pivotal role in the detection and analysis of epileptic seizures, which affects over 70 million people in the world. Nonetheless, the visual interpretation of EEG signals for epilepsy detection is laborious and time-consuming. To tackle this open challenge, we introduce a straightforward yet efficient hybrid deep learning approach, named ResBiLSTM, for detecting epileptic seizures using EEG signals. Firstly, a one-dimensional residual neural network (ResNet) is tailored to adeptly extract the local spatial features of EEG signals. Subsequently, the acquired features are input into a bidirectional long short-term memory (BiLSTM) layer to model temporal dependencies. These output features are further processed through two fully connected layers to achieve the final epileptic seizure detection. The performance of ResBiLSTM is assessed on the epileptic seizure datasets provided by the University of Bonn and Temple University Hospital (TUH). The ResBiLSTM model achieves epileptic seizure detection accuracy rates of 98.88–100% in binary and ternary classifications on the Bonn dataset. Experimental outcomes for seizure recognition across seven epilepsy seizure types on the TUH seizure corpus (TUSZ) dataset indicate that the ResBiLSTM model attains a classification accuracy of 95.03% and a weighted F1 score of 95.03% with 10-fold cross-validation. These findings illustrate that ResBiLSTM outperforms several recent deep learning state-of-the-art approaches.

show abstract

Section: Discussionmentioning

confidence: 79%

Section: Discussionmentioning

confidence: 79%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Residual and bidirectional LSTM for epileptic seizure detection

Zhao,

Wang,

Patnaik

et al. 2024

Front. Comput. Neurosci.

View full text Add to dashboard Cite

show abstract

“…Due to this drawback in LSTM, the concept of decision fusion, which combines DL algorithm outputs, is applied for precise and accurate classification prediction and has obtained improved performance [33] with increased computational cost. Previous studies using LSTM in combination with CNN and other DL neural networks [40,41], where CNN and LSTM are combined to extract the features from complex brain patterns for more precise classification, obtained enhanced performance of fusion of DL models at the cost of increased processing and computational time. Fernandez Rojas et al [42] present a hybrid CNN-LSTM model with an accuracy of 91.2 ± 11.7, compared to 86.4 ± 16.8 and 88.4 ± 21.1 for the CNN and LSTM models, respectively.…”

Section: Discussionmentioning

confidence: 99%

Enhancing Classification Accuracy with Integrated Contextual Gate Network: Deep Learning Approach for Functional Near-Infrared Spectroscopy Brain–Computer Interface Application

Akhter,

Naseer,

Nazeer

et al. 2024

Sensors

View full text Add to dashboard Cite

Brain–computer interface (BCI) systems include signal acquisition, preprocessing, feature extraction, classification, and an application phase. In fNIRS-BCI systems, deep learning (DL) algorithms play a crucial role in enhancing accuracy. Unlike traditional machine learning (ML) classifiers, DL algorithms eliminate the need for manual feature extraction. DL neural networks automatically extract hidden patterns/features within a dataset to classify the data. In this study, a hand-gripping (closing and opening) two-class motor activity dataset from twenty healthy participants is acquired, and an integrated contextual gate network (ICGN) algorithm (proposed) is applied to that dataset to enhance the classification accuracy. The proposed algorithm extracts the features from the filtered data and generates the patterns based on the information from the previous cells within the network. Accordingly, classification is performed based on the similar generated patterns within the dataset. The accuracy of the proposed algorithm is compared with the long short-term memory (LSTM) and bidirectional long short-term memory (Bi-LSTM). The proposed ICGN algorithm yielded a classification accuracy of 91.23 ± 1.60%, which is significantly (p < 0.025) higher than the 84.89 ± 3.91 and 88.82 ± 1.96 achieved by LSTM and Bi-LSTM, respectively. An open access, three-class (right- and left-hand finger tapping and dominant foot tapping) dataset of 30 subjects is used to validate the proposed algorithm. The results show that ICGN can be efficiently used for the classification of two- and three-class problems in fNIRS-based BCI applications.

show abstract