Pulmonary Nodule Classification with Deep Convolutional Neural Networks on Computed Tomography Images

Figure 2

Architecture of our CNN for nodule recognition. The input data is ROI image pixels as a 1024-dimensional vector, and the number of output neurons of the network is 2 (nodule: 1 and nonnodule: 0). The numbers of neurons in the other layers are set to 6272, 1568, 1600, 250, 150, 100, and 50.