Skip to main content

Table 6 Generalizability results of the models trained on Kvasir-SEG and tested on CVC-Clinicdb

From: EG-TransUNet: a transformer-based U-Net with enhanced and guided models for biomedical image segmentation

Method

Year

mDice

mIoU

Recall

Precision

U-Net [5]

2015

0.6302

0.5015

0.5612

0.8249

U-Net++ [7]

2018

0.4267

0.3623

0.4337

0.6877

HRNetV2-W18-Smallv2 [44]

2020

0.6428

0.5513

0.6811

0.7253

HRNetV2-W48 [44]

2020

0.7901

0.6953

0.8796

0.7694

MSRF-Net [41]

2021

0.7921

0.6498

0.9001

0.7694

EG-TransUNet

–

0.8939

0.8420

0.9020

0.9147