Table 5 The performance of different Student Segments under the same Teacher Annotator on LUNA16 dataset.

From: A semisupervised knowledge distillation model for lung nodule segmentation

Student

Teacher

ACC

DICE

Sens

Spec

HF95

UTransformer

UNet

0.9665

0.3920

0.5445

0.9750

27.45

AttUNet

UNet

0.9705

0.3783

0.6122

0.9868

12.40

ResUNet

UNet

0.9932

0.4573

0.7389

0.9850

16.20

SwinUNet

UNet

0.9806

0.4623

0.7401

0.9855

11.78

SSLKD-UNet

UNet

0.9850

0.5253

0.7603

0.9931

8.50

  1. The Student column represents the model used by the Student Segmentor. The bold result represents the best result among the relevant indicators.
Back to article page