Figure 5
From: Automatic classification of normal and abnormal cell division using deep learning
Diagram of the network architectures. For each mitosis video frame, it goes through the same set of convolutional layers (Xception, VGG16, Resnet50 or Gao’s architectures) and is reduced to a feature vector. These vectors are then input to an LSTM layer to capture the temporal relationships and produce an output vector. The output vector goes through FC layers and finally yields a normal or abnormal classification.
