Figure 3 | Scientific Reports

Figure 3

From: Automated recognition of epilepsy from EEG signals using a combining space–time algorithm of CNN-LSTM

Figure 3The alternative text for this image may have been generated using AI.

Pseudocode of CNN algorithm. where \(u_{m}^{l}\) represent the \(m^{th}\) channel activation value of the \(1^{th}\) convolutional layer, \(u_{m}^{l}\) is obtained by the convolution of the upper layer and bias, \(x_{m}^{l}\) is the \(m^{th}\) channel output of \(1^{th}\) convolution layer, \(f( \cdot )\) is the activation function, \(p\) is the selected feature sets of input signals, \(H_{mn}^{l}\) represents a convolution function, \(b_{m}^{l}\) is the bias of \(u_{m}^{l}\), \(down( \cdot )\) represents the \(down\) sample function, \(a_{m}^{l}\) is the offset coefficient, \(a_{m}^{1}\), \(b_{m}^{1}\) are the bias coefficient of the feature, \(u^{l}\) is the activation value of the \(l^{th}\) fully-connected layer, \(w^{l}\) represent the weight of the fully-connected layer, and \(b^{l}\) are bias.

Back to article page