Researchers have developed MAX (MAnifold eXpansion), a novel subject-specific framework designed to improve accelerated multi-contrast MRI reconstruction. MAX learns anatomical representations from a single fully sampled reference contrast, expanding the multi-contrast manifold through anatomy-preserving intensity augmentations. This approach achieves superior performance, outperforming baselines by over 1 dB in PSNR for brain and knee MRI reconstructions at acceleration factors of R=8 and R=6, respectively. The framework demonstrates robustness to motion, structural heterogeneity, and noise, offering a general strategy for leveraging reference scans in accelerated MRI. AI
IMPACT Improves medical imaging accuracy and efficiency, potentially leading to better diagnoses and faster scans.
RANK_REASON Research paper detailing a new method for medical imaging reconstruction. [lever_c_demoted from research: ic=1 ai=0.7]
Read on arXiv cs.IR (Information Retrieval) →
- alphaXiv
- arXiv
- CatalyzeX
- DagsHub
- Gotit.pub
- Hugging Face
- magnetic resonance imaging
- MAnifold eXpansion
- MAX
- peak signal-to-noise ratio
- ScienceCast
- Structural Similarity Index Measure
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
