Fig. 4.
Architecture of the CNN model. The CNN architecture comprises three layers: two consecutive convolutional-pooling layers and a fully connected classification layer. The two convolutional-pooling layers use the same fixed convolutional kernel and pooling kernel, but have 64 and 128 neurons, respectively. The last layer has 256 neurons, which are all connected to the final two neurons for mitosis/nonmitosis classification.
