Skip to main content
. 2022 Aug 5;12:13460. doi: 10.1038/s41598-022-17341-6

Figure 4.

Figure 4

The 3D reconstructed whole canal from the ground truth and segmentation results by Canal-Net (ours), ConvLSTM 3D U-Net (ours), MPL 3D U-Net (ours), 3D U-Net, SegNet, and 2D U-Net from the top to bottom (a–f) from the same canals shown in Fig. 3a–f. The red arrows indicate the position of the corresponding image slice in Fig. 3.