Figure 4
From: Bladder Cancer Treatment Response Assessment in CT using Radiomics with Deep-Learning
DL-CNN Structure. An input ROI is convolved with multiple convolution kernels, and the resulting values are collected into the corresponding kernel maps. This process repeats for several layers, giving the “deep” convolutional neural network. The network used in this study contains two convolution layers and two locally-connected layers, each of which contains 16 kernels.
