Research map / Image generation / Image restoration
Data augmentation for deep learning based accelerated MRI reconstruction with limited data
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 |
|---|---|---|---|
| Apply MRAugment instead of no data augmentation | E2E VarNet · Stanford 2D FSE, full training set | SSIM: 0.8950 -> 0.9120 | Helped |
| Apply physics-informed data augmentation | E2E VarNet · fastMRI knee, 2% train, unseen scanner transfer | SSIM: 0.8646 -> 0.9049; 0.8241 -> 0.8551; 0.8174 -> 0.8913 | Helped |
| Apply physics-informed data augmentation | E2E VarNet · fastMRI knee, 100% train, unseen scanner transfer | SSIM: 0.9177 -> 0.9185; 0.8686 -> 0.8690; 0.9043 -> 0.9062 | Helped |
| Use pixel-preserving transformations only (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8396 -> 0.8585 | Helped |
| Use interpolating transformations only (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8396 -> 0.8731 | Helped |
| Combine pixel-preserving and interpolating transformations (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8731 -> 0.8758 | Helped |
| Use exponential augmentation schedule with p=0.3 (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8396 -> 0.8565 | Helped |
| Use constant augmentation probability p=0.3 (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8396 -> 0.8588 | Helped |
| Use exponential rather than constant schedule at p=0.3 (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8588 -> 0.8565 | Hurt |
| Use exponential augmentation schedule with p=0.6 (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8396 -> 0.8758 | Helped |
| Use constant augmentation probability p=0.6 (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8396 -> 0.8611 | Helped |
| Increase exponential schedule probability to p=0.8 (ablation) | MRAugment · fastMRI knee, 1% train, multi-coil | SSIM: 0.8758 -> 0.8600 | Hurt |
Closest papers by meaning
- Improving Deep Learning for Accelerated MRI With Data Filtering (NeurIPS 2025 · similarity 0.91)
- Robustness of Deep Learning for Accelerated MRI: Benefits of Diverse Training Data (ICML 2024 · similarity 0.89)
- Efficient Noise Calculation in Deep Learning-based MRI Reconstructions (ICML 2025 · similarity 0.88)
- A Unified Model for Compressed Sensing MRI Across Undersampling Patterns (CVPR 2025 · similarity 0.85)
- UGoDIT: Unsupervised Group Deep Image Prior Via Transferable Weights (NeurIPS 2025 · similarity 0.85)
- D2SA: Dual-Stage Distribution and Slice Adaptation for Efficient Test-Time Adaptation in MRI Reconstruction (NeurIPS 2025 · similarity 0.85)
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.