Skip to main content
Fig. 5 | BMC Medical Imaging

Fig. 5

From: Development and validation of the 3D U-Net algorithm for segmentation of pelvic lymph nodes on diffusion-weighted images

Fig. 5

Examples of the detection results of the 3D U-Net for the lymph nodes. The first column represents the DWI images; the second column represents the manual annotations, and the third column represents the automated segmentations. a The true positive LNs detected by the model (Green label in automated segmentation); b The false positive detection that was a part of the colon (Green label in automated segmentation); c The false negative detection that was missed by the model (Arrow points in the automated segmentation)

Back to article page

BMC Medical Imaging

ISSN: 1471-2342

Contact us