Fig 1.
Axial slice of a ground truth pancreas segmentation in an abdominal CT scan (MSD), cropped to show detail of surrounding tissues.
Fig 2.
Schematic representation of the MoNet architecture.
Fig 3.
Schematic representation diagram of a RDDC block (top) and the constituent convolutional (bottom).
Fig 4.
Exemplary segmentation results (yellow) of: U-Net-16 (A), Attention U-Net (B), U-Net-64 (C), MoNet (D), on the pancreas MSD validation set, Ground truth indicated by red outline. Box-plots of Hausdorff distances (E) and Dice scores (F) computed for the whole pancreas MSD validation set on a per-patient basis.
Table 1.
Comparison of MoNet with other U-Net variants in two different imaging modalities on the task of pancreas and brain lesion segmentation, CT and MRI respectively.
We report performance on validation sets of the MSD datasets (brain tumor and pancreas) as well as out-of sample generalization performance on an independent validation set (IVD), collected and annotated in-house.
Table 2.
CPU inference time (sec) for a CT scan of 150 slices and timer per batch (sec) on GPU, both at a resolution of 256 × 256.
Table 3.
Comparison of storage space occupied by MoNet and other U-Net variants.
Fig 5.
Input image (A) and target (B) alongside visualizations of the first 16 channels of intermediate activations for the given input image produced by early (1), middle (2) and late (3) convolutional layers in U-net (C) and MoNet (D). Histograms computed for all channels in the feature maps for early (1), middle (2), and late (3) convolution layers for U-net (E) and MoNet (F).