Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
image
image
label
class label
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
0images
End of preview. Expand in Data Studio

MedCAGD: Context-Aware Gated Decoder for Efficient Medical Image Segmentation (ECCV 2026) - Medical Image Segmentation Datasets

This repo contains multiple publicly available medical image segmentation (Semantic Segmentation) datasets used for training and evaluation of the segmentation models. Each subfolder corresponds to a specific dataset and follows its original structure or a standardized format used in this project. These datasets are used for benchmarking segmentation performance across multiple medical imaging modalities including CT, MRI, dermoscopy, endoscopy, ultrasound, fundus imaging, and microscopy.

The segmentation benchmarks included in this dataset collection correspond to the evaluations reported in the paper (MedCAGD). Detailed benchmark results, evaluation protocols, and methodological descriptions are available in the publication.

Citation

If you use this, please cite the following papers

@inproceedings{wazir2026medcagdcontextawaregateddecoder,
      title={MedCAGD: Context-Aware Gated Decoder for Efficient Medical Image Segmentation}, 
      author={Saad Wazir, Patrick Dominique Vibild, Dinh Phu Tran, Seongah Kim, Daeyoung Kim},
      year={2026},
      eprint={2607.00409},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2607.00409}, 
}
@inproceedings{wazir2025rethinking,
  title={Rethinking decoder design: Improving biomarker segmentation using depth-to-space restoration and residual linear attention},
  author={Wazir, Saad and Kim, Daeyoung},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={30861--30871},
  year={2025},
  doi = {10.48550/arXiv.2506.18335},
  url = {https://doi.org/10.48550/arXiv.2506.18335}
}

MedCAGD: Context-Aware Gated Decoder for Efficient Medical Image Segmentation - ECCV 2026

GitHub | Paper

MCADS-Decoder - Rethinking decoder design: Improving Biomarker Segmentation Using Depth-to-Space Restoration and Residual Linear Attention - CVPR 2025

GitHub | Paper


MedCAGD-Dataset-Viewer

https://huggingface.co/spaces/saadwazir/MedCAGD-Dataset-Viewer

Included Datasets:

ACDC-2D-Slices 2D cardiac MRI slices from the ACDC dataset used for cardiac structure segmentation.

Synapse Multi-Organ Segmentation Dataset-8 abdominal organs-2D-Slices
2D slices from the Synapse multi-organ CT segmentation dataset containing annotations for eight abdominal organs.

ThyroidXL
Thyroid ultrasound dataset for thyroid nodule segmentation.

ISIC2017
Skin lesion segmentation dataset from the ISIC 2017 challenge.

ISIC2018
Skin lesion segmentation dataset from the ISIC 2018 challenge.

BKAI
Gastrointestinal polyp segmentation dataset released by BKAI.

ClinicDB
Colonoscopy polyp segmentation dataset from the CVC-ClinicDB benchmark.

ColonDB
Colonoscopy polyp segmentation dataset commonly used for evaluating polyp detection models.

ETIS
ETIS-Larib polyp dataset containing challenging colonoscopy images with pixel-level annotations.

Kvasir
Kvasir-SEG dataset for gastrointestinal polyp segmentation.

BUSI
Breast ultrasound dataset used for tumor segmentation.

DRIVE
Retinal vessel segmentation dataset from the DRIVE challenge.

FIVES
Fundus Image Vessel Segmentation dataset for retinal blood vessel analysis.

CHASEDB Retinal vessel segmentation dataset from the CHASE_DB1 benchmark.

LES-AV
Dataset for retinal artery and vein segmentation.

STARE
Retinal vessel segmentation dataset from the STARE benchmark.

UOA-DR
Diabetic retinopathy dataset with retinal lesion annotations.

CellSeg
Microscopy cell segmentation dataset used for cell instance or semantic segmentation.


Benchmark Results

Table: 1 - Comprehensive performance comparison across 9 medical image segmentation benchmarks. Average Dice scores are reported.
Method Params ↓ FLOPs ↓ Skin Polyp Fundus Neoplasm Cell All
ISIC17ISIC18 ETISColonDB DRIVEFIVES BUSIThyroidXL CellSegAvg
U-Net34.53 M65.53 G83.0786.6776.8583.9571.2075.7774.0471.1671.5277.14
AttnUNet34.88 M66.64 G83.6687.0576.8486.4671.6875.9974.4872.5072.6477.92
DeepLabv3+39.76 M14.92 G83.8488.6490.7391.9269.5975.1276.8173.4671.9080.22
UNet++09.16 M34.65 G82.9887.4677.4087.8872.9485.7474.4683.9478.3081.23
nnU-Net31.29 M55.26 G83.2388.5380.1391.6375.4376.1076.4686.0883.5382.34
PraNet32.55 M06.93 G83.0388.5683.8489.1675.2184.5775.1485.5179.0782.68
TransUNet105.32 M38.52 G85.0089.1687.7991.6374.9883.5478.3085.7779.0883.92
Swin-Unet27.17 M06.20 G83.9789.2685.1089.2774.9384.1777.3885.8078.8483.19
UCTransNet65.60 M56.70 G83.2789.1887.3591.6575.4284.7479.5385.8279.3384.03
UNeXt1.470 M0.570 G82.7487.7874.0383.8474.7776.6074.7184.4675.7179.40
VM-UNet27.43 M04.12 G85.9987.0585.5288.7173.2583.5174.6978.3174.9481.33
Swin-UMamba60.00 M68.00 G83.4087.6286.6387.9773.3282.6673.3884.9675.5681.72
EMCAD26.76 M05.60 G85.9590.9692.2992.3177.1582.5180.2583.3379.1384.87
MCADS50.90 M61.89 G84.1491.0192.2491.3778.4276.0580.0386.3386.6885.14
Ours30.60 M05.00 G86.6191.5693.4793.2781.6387.5083.4788.0286.6188.01
AutoSam*41.56 M25.11 G--79.7083.00------
Medical SAM3*840.0 M---86.10-55.80-----
Table: 2 - Performance comparison with SOTA methods on the Synapse multi-organ dataset.
MethodDice ↑IoU ↑HD95 ↓AortaGBKLKRLiverPCSPSM
U-Net70.1159.3944.6984.0056.7072.4162.6486.9848.7381.4867.96
AttnUNet71.7068.0926.0184.0466.4257.2684.5381.2873.8766.0660.17
UNet++72.3968.8225.6183.6567.6657.2684.5381.3473.8768.9761.85
nnU-Net75.3371.4719.3477.0673.2776.3484.5379.9873.3477.6260.52
PraNetV283.7574.8117.7788.6972.7985.4182.9195.8268.4793.0985.85
TransUNet77.6167.3226.9086.5660.4380.5478.5394.3358.4787.0675.00
Swin-Unet77.5866.8827.3281.7665.9582.3279.2293.7353.8188.0475.79
UCTransNet79.0875.4115.5983.0681.3577.2478.2385.7674.7781.8970.31
UNETR*78.35-18.5989.8056.3085.6084.5294.5760.4785.0070.46
MISSFormer*81.96-18.2086.9968.6585.2182.0094.4165.6791.9280.81
U-Mamba78.6374.8716.1983.7778.7079.4082.3783.8674.7879.7766.41
VM-UNet73.3971.6127.9763.5772.6277.9892.5979.4470.8055.5874.55
EMCAD83.6374.6515.6888.1468.8788.0884.1095.2668.5192.1783.92
MCADS85.0381.7111.1190.8186.0786.7783.2487.6683.5585.7476.38
Ours87.00±0.283.7714.3992.2890.3189.7287.2191.0282.0886.9176.51
Self-Prompt SAM*86.74--91.9969.9585.6585.4097.3979.1894.3889.94
Table: 3 - Performance comparison with SOTA methods on the ACDC dataset.
MethodDice ↑IoU ↑HD95 ↓RVMyoLV
U-Net81.5673.416.985476.9980.2887.43
AttnUNet82.3773.946.168478.1381.0887.89
UNet++81.9773.926.472477.7480.7387.44
nnU-Net82.6674.276.166379.0081.0187.97
PraNetV283.7476.136.371979.6183.1088.51
TransUNet83.0774.855.757879.1681.6588.41
Swin-Unet82.6174.596.124478.9480.1788.73
UCTransNet84.8977.575.699580.9484.1189.62
U-Mamba84.1876.475.850180.9083.2488.40
VM-UNet81.0272.747.002576.7579.4086.90
EMCAD85.0777.735.247281.5884.2389.42
MCADS84.5176.925.559581.1683.2789.09
Ours87.54±0.380.964.405785.2786.2391.11

Acknowledgement

We gratefully acknowledge the prior contributions of the research community, which have provided the foundation for our framework.

Downloads last month
295

Space using saadwazir/MedCAGD-Dataset-Collection 1

Papers for saadwazir/MedCAGD-Dataset-Collection

Free AI Image Generator No sign-up. Instant results. Open Now