A Semiautomated Deep Learning Approach for Pancreas Segmentation

Table 1

Quantitative comparisons between the three different models on the NIH dataset. Bold denotes the best.

Model	F-measure	Recall	Precision	Mean DSC
Modified Deformable-ConvNet	0.8201	0.8084	0.8378	0.8203
U-Net	0.8738	0.9010	0.8499	0.8670
DUNet(Ours)	0.8878	0.8997	0.8898	0.8725