Fig. 4
From: Multi-view hybrid encoder U-Net for 3D renal vascular medical image segmentation
Visual comparison of segmentation results for Kidney_3 from three orthogonal views. Subfigures (a–c) correspond to the axial (XY), coronal (XZ), and sagittal (YZ) views, respectively. In each row, columns (1), (2), and (3) represent: (1) the ground truth overlaid with the kidney organ, (2) the predicted result overlaid with the kidney organ, and (3) the overlay of ground truth and predicted result with the kidney organ removed. This visualization highlights the consistency and accuracy of the proposed method across different anatomical planes.
