DDR Dataset
A public medical image dataset for lesion segmentation in diabetic retinopathy, containing annotations for multiple lesion types such as microaneurysms, hemorrhages, and hard exudates.
Timeline (last 90 days)
The D²Mamba model achieved 33.87% mIoU in five-fold cross-validation on the DDR dataset, 2.99 percentage points higher than the strongest baseline.
DDR (Diabetic Retinopathy Dataset) is a publicly available dataset widely used in diabetic retinopathy analysis, containing color fundus photographs and pixel-level lesion annotations, covering lesion types such as microaneurysms, hemorrhages, hard exudates, and soft exudates.
This study has so far completed five-fold cross-validation on only a single dataset (DDR); cross-dataset generalization ability and stability under different imaging devices still require further testing.
All Facts (3)
The D²Mamba model achieved 33.87% mIoU in five-fold cross-validation on the DDR dataset, 2.99 percentage points higher than the strongest baseline.
50%UnverifiedDDR (Diabetic Retinopathy Dataset) is a publicly available dataset widely used in diabetic retinopathy analysis, containing color fundus photographs and pixel-level lesion annotations, covering lesion types such as microaneurysms, hemorrhages, hard exudates, and soft exudates.
50%UnverifiedThis study has so far completed five-fold cross-validation on only a single dataset (DDR); cross-dataset generalization ability and stability under different imaging devices still require further testing.
50%