Medical Image Segmentation On Etis

Metrics

mIoU

mean Dice

Results

Performance results of various models on this benchmark

Model Name	mIoU	mean Dice	Paper Title	Repository
MEGANet(ResNet-34)	0.709	0.789	MEGANet: Multi-Scale Edge-Guided Attention Network for Weak Boundary Polyp Segmentation	-
DUCK-Net	0.8788	0.9354	Using DUCK-Net for Polyp Image Segmentation	-
RSAFormer	-	0.835	RSAFormer: A method of polyp segmentation with region self-attention transformer
DuAT	0.746	0.822	DuAT: Dual-Aggregation Transformer Network for Medical Image Segmentation	-
SSFormer-L	0.720	0.796	Stepwise Feature Fusion: Local Guides Global	-
ESFPNet-L	0.748	0.823	ESFPNet: efficient deep learning architecture for real-time lesion segmentation in autofluorescence bronchoscopic video	-
EMCAD	-	0.9229	EMCAD: Efficient Multi-scale Convolutional Attention Decoding for Medical Image Segmentation	-
Meta-Polyp	0.704	0.78	Meta-Polyp: a baseline for efficient Polyp segmentation	-
ProMISe	0.750	0.840	ProMISe: Promptable Medical Image Segmentation using SAM	-
UACANet-L	0.689	0.766	UACANet: Uncertainty Augmented Context Attention for Polyp Segmentation	-
PVT-CASCADE	0.7258	0.8007	Medical Image Segmentation via Cascaded Attention Decoding
ResUNet++	0.7534	0.6364	ResUNet++: An Advanced Architecture for Medical Image Segmentation	-
PraNet	0.5670	0.6280	PraNet: Parallel Reverse Attention Network for Polyp Segmentation	-
MEGANet(Res2Net-50)	0.665	0.739	MEGANet: Multi-Scale Edge-Guided Attention Network for Weak Boundary Polyp Segmentation	-
HarDNet-MSEG	0.613	0.677	HarDNet-MSEG: A Simple Encoder-Decoder Polyp Segmentation Neural Network that Achieves over 0.9 Mean Dice and 86 FPS	-
COMMA (Res2Net-50)	0.648	0.711	COMMA: Propagating Complementary Multi-Level Aggregation Network for Polyp Segmentation	-
RAPUNet	0.9179	0.9572	MetaFormer and CNN Hybrid Model for Polyp Image Segmentation
TransFuse-L	0.661	0.737	TransFuse: Fusing Transformers and CNNs for Medical Image Segmentation	-
TransFuse-S	0.659	0.733	TransFuse: Fusing Transformers and CNNs for Medical Image Segmentation	-
HarDNet-DFUS	-	0.730	HarDNet-DFUS: An Enhanced Harmonically-Connected Network for Diabetic Foot Ulcer Image Segmentation and Colonoscopy Polyp Segmentation	-

0 of 24 row(s) selected.