Lesion Segmentation
Lesion segmentation is a critical task in medical image processing, aiming to accurately separate the lesion area from other tissue structures. This technique achieves automatic recognition and boundary demarcation of lesions through advanced algorithms, providing key support for clinical diagnosis, treatment planning, and efficacy evaluation.
Anatomical Tracings of Lesions After Stroke (ATLAS)
MGA-Net
Anatomical Tracings of Lesions After Stroke (ATLAS)
D-UNet
BUS 2017 Dataset B
Attn U-Net + Multi-Input + FTL
HAM10000
DermoSegDiff-B
ISIC 2017
Automatic skin lesion segmentation with fully convolutional-deconvolutional networks
ISIC 2018
Polar Res-U-Net++
ISIC 2018 Task 1
DCSAU-Net
ISLES-2015
3D CNN + CRF
PH2
University of Waterloo skin cancer database
DTP-Net