Medical Image Segmentation On Synapse Multi

Metrics

Avg DSC

Avg HD

Results

Performance results of various models on this benchmark

Model Name	Avg DSC	Avg HD	Paper Title	Repository
AgileFormer	86.11	12.88	AgileFormer: Spatially Agile Transformer UNet for Medical Image Segmentation
PAG-TransYnet	83.43	15.82	Rethinking Attention Gated with Hybrid Dual Pyramid Transformer-CNN for Generalized Segmentation in Medical Imaging	-
nnFormer	86.57	10.63	nnFormer: Interleaved Transformer for Volumetric Segmentation	-
MedSegDiff-v2	89.50	-	MedSegDiff-V2: Diffusion based Medical Image Segmentation with Transformer
nnUNet	88.80	10.78	nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation
SETR	79.60	-	Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers
ParaTransCNN	83.86	15.86	ParaTransCNN: Parallelized TransCNN Encoder for Medical Image Segmentation
EMCAD	83.63	15.68	EMCAD: Efficient Multi-scale Convolutional Attention Decoding for Medical Image Segmentation
UCTransNet	78.99	30.29	UCTransNet: Rethinking the Skip Connections in U-Net from a Channel-wise Perspective with Transformer
Interactive AI-SAM gt box	90.66	-	AI-SAM: Automatic and Interactive Segment Anything Model
SegFormer3D	82.15	-	SegFormer3D: an Efficient Transformer for 3D Medical Image Segmentation	-
TransUNet	81.19	-	S2S2: Semantic Stacking for Robust Semantic Segmentation in Medical Imaging
MERIT	84.90	13.22	Multi-scale Hierarchical Vision Transformer with Cascaded Attention Decoding for Medical Image Segmentation
TransUNet	77.48	31.69	TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation
Medical SAM Adapter	89.80	-	Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentation
MIST	86.92	11.07	MIST: Medical Image Segmentation Transformer with Convolutional Attention Mixing (CAM) Decoder
Automatic AI-SAM	84.21	-	AI-SAM: Automatic and Interactive Segment Anything Model
MISSFormer	81.96	18.20	MISSFormer: An Effective Medical Image Segmentation Transformer
MedNeXt-L (5x5x5)	88.76	-	MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentation
FCB Former	80.26	-	Adaptive t-vMF Dice Loss for Multi-class Medical Image Segmentation

0 of 23 row(s) selected.