Fig. 2
From: Quality assurance for automatically generated contours with additional deep learning
Network architecture. An EfficientNet B0 backbone is connected to three repeated bidirectional feature pyramid blocks (BiFPNs). The regression head consists of serially connected fast normalized fusion nodes and finally BatchNorm (BN), PReLU, a single-channel convolution, and a sigmoid activation. Numbers indicate the resolution at each level relative to the input. Image adapted from [35]