Research map / Image generation / Super-resolution
Deep Multi-Task Learning for Diabetic Retinopathy Grading in Fundus Images
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use DeepMT-DR instead of the strongest listed baseline | DR grading · DDR | accuracy: 79.5 -> 82.5 | Helped |
| Use DeepMT-DR instead of the strongest listed baseline | DR grading · EyePACS | accuracy: 83.1 -> 85.7 | Helped |
| Jointly train the ISR, lesion segmentation, and grading tasks rather than remove ISR or train it independently (ablation) | DeepMT-DR · DDR | DR grading performance | Helped |
| Jointly train lesion segmentation with the other tasks rather than remove it or train it independently (ablation) | DeepMT-DR · DDR | DR grading performance | Helped |
| Add segmentation-aware loss (ablation) | DeepMT-DR · DDR | DR grading performance | Helped |
| Add classification-aware loss (ablation) | DeepMT-DR · DDR | DR grading performance | Helped |
| Use GMSV instead of GBP for network visualization (ablation) | DeepMT-DR · DDR | accuracy: 81.4 -> 82.5 | Helped |
| Use GMSV instead of Grad-CAM for network visualization (ablation) | DeepMT-DR · DDR | accuracy: 81.5 -> 82.5 | Helped |
| Use GMSV instead of SmoothBP for network visualization (ablation) | DeepMT-DR · DDR | accuracy: 81.7 -> 82.5 | Helped |
| Use DeepMT-DR rather than MDRN for super-resolution | ISR · DDR | PSNR: 38.9 -> 39.7 | Helped |
| Use DeepMT-DR rather than MDRN for super-resolution | ISR · DDR | SSIM: 0.856 -> 0.883 | Helped |
Closest papers by meaning
- HACDR-Net: Heterogeneous-Aware Convolutional Network for Diabetic Retinopathy Multi-Lesion Segmentation (AAAI 2024 · similarity 0.90)
- Like an Ophthalmologist: Dynamic Selection Driven Multi-View Learning for Diabetic Retinopathy Grading (AAAI 2025 · similarity 0.89)
- Deep Hierarchies and Invariant Disease-Indicative Feature Learning for Computer Aided Diagnosis of Multiple Fundus Diseases (AAAI 2025 · similarity 0.87)
- FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs (NeurIPS 2026 · similarity 0.86)
- SIGraph: Saliency Image-Graph Network for Retinal Disease Classification in Fundus Image (AAAI 2025 · similarity 0.84)
- Domain Adaptive Diabetic Retinopathy Grading with Model Absence and Flowing Data (CVPR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.