August 25, 2020Open Access

Identification of post-stroke EEG signal using wavelet and convolutional neural networks

Q: What is the clinical evidence from this study?

Study design: Other. Population: Post-stroke (n=75). Intervention: Convolutional Neural Network (CNN) with wavelet extraction (amplitude and Beta features) vs. CNN without amplitude or Beta features. Primary outcome: Classification accuracy on testing data.

Resultado clave

A convolutional neural network using wavelet extraction with amplitude and Beta wave features achieved 90% accuracy in classifying post-stroke EEG signals compared to 70% without these features.

PICO estructurado

Población

75 subjects, comprising 50 post-stroke outpatients and 25 healthy volunteers, whose EEG signals were recorded to classify stroke versus no stroke.

Intervención

EEG signal classification using Convolutional Neural Networks (CNN) with wavelet extraction (incorporating Delta, Alpha, Beta, Theta, Mu waves, and amplitude features).

Comparador

Alternative feature configurations (without amplitude or Beta waves) and optimization models (Adam vs. Stochastic Gradient Descent [SGD]).

Resultado

Accuracy of identifying post-stroke patients ('minor stroke' vs 'no stroke') from EEG signals.

A Convolutional Neural Network model using wavelet extraction of EEG signals, specifically incorporating amplitude and Beta wave features, can identify post-stroke patients with 90% accuracy.

Resultado numérico

Tasa de eventos absoluta: 90% vs 70%

Limitaciones

Need to find the optimal configuration of EEG channels to increase accuracy and reduce computing time
Setting and sequence of features require further attention to balance high accuracy with low computing time

Resumen

Post-stroke patients need ongoing rehabilitation to restore dysfunction caused by an attack so that a monitoring device is required. EEG signals reflect electrical activity in the brain, which also informs the condition of post-stroke patient recovery. However, the EEG signal processing model needs to provide information on the post-stroke state. The development of deep learning allows it to be applied to the identification of post-stroke patients. This study proposed a method for identifying post-stroke patients using convolutional neural networks (CNN). Wavelet is used for EEG signal information extraction as a feature of machine learning, which reflects the condition of post-stroke patients. This feature is Delta, Alpha, Beta, Theta, and Mu waves. Moreover, the five waves, amplitude features are also added according to the characteristics of the post-stroke EEG signal. The results showed that the feature configuration is essential as distinguish. The accuracy of the testing data was 90% with amplitude and Beta features compared to 70% without amplitude or Beta. The experimental results also showed that adaptive moment estimation (Adam) optimization model was more stable compared to Stochastic gradient descent (SGD). But SGD can provide higher accuracy than the Adam model.

Me gusta

Guardar

Ver artículo completo

Me gusta

Guardar

Ver artículo completo

Cite This Study

Djamal et al. (Tue,) conducted a other in Post-stroke (n=75). Convolutional Neural Network (CNN) with wavelet extraction (amplitude and Beta features) vs. CNN without amplitude or Beta features was evaluated on Classification accuracy on testing data. A convolutional neural network using wavelet extraction with amplitude and Beta wave features achieved 90% accuracy in classifying post-stroke EEG signals compared to 70% without these features.

synapsesocial.com/papers/6a2ac78be5f3e3038ed7c7f4 https://doi.org/https://doi.org/10.11591/eei.v9i5.2005