Skip to main content

Table 2 Performance of different input dimensions

From: Development and evaluation of a deep learning framework for pelvic and sacral tumor segmentation from multi-sequence MRI: a retrospective study

Input dimension

Model

Dice Scorea

IoUb

3D

nnUnet

0.639

0.509

2D

MobilenetV2 + Unet

0.693

0.560

2.5D

MobilenetV2 + Unet

0.741

0.615

  1. Note:
  2. a, Dice score is 2×the area of overlap divided by the total number of pixels in both images
  3. b, IoU is the area of overlap divided by the area of union of both images
Back to article page

Cancer Imaging

ISSN: 1470-7330

Contact us