Research map / Vision-language and multimodal / Cross-modal learning
Virtual Full-stack Scanning of Brain MRI via Imputing Any Quantised Code
Yicheng Wu, Tao Song, Zhonghua Wu, Jin Ye, Zongyuan Ge, Wenjia Bai, Zhaolin Chen, Jianfei Cai
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add common feature map Fc to the Stage I decoder (ablation) | CodeBrain · IXI reconstruction | PSNR (dB): 30.15 -> 34.32 | Improved |
| use grading (ordinal) loss instead of classification for code prediction (ablation) | CodeBrain · IXI imputation mean | PSNR (dB): 29.24 -> 29.50 | Improved |
| use PSNR loss instead of L1 loss for reconstruction (ablation) | CodeBrain · IXI reconstruction | PSNR / SSIM: PSNR 34.08 -> 34.32; SSIM 96.28 -> 95.94 | Mixed |
| use lambda_m 20, lambda_a 5 rather than other weights (ablation) | CodeBrain · IXI reconstruction | PSNR (dB): 33.83 -> 34.32 | Improved |
| use region-level quantised codes as conditions instead of fixed binary conditions (ablation) | CodeBrain · IXI | PSNR (dB): 28.27 -> 29.50 | Improved |
| use continuous conditions instead of quantised codes (ablation) | CodeBrain · IXI imputation | PSNR (dB): 29.50 -> 28.87 | Hurt |
| use CodeBrain instead of Zhang et al. | Zhang et al. · IXI mean | PSNR (dB): 29.00 -> 29.50 | Improved |
| use CodeBrain instead of Zhang et al. on BraTS | Zhang et al. · BraTS 2023 mean | PSNR (dB): 25.01 -> 25.31 | Improved |
| use CodeBrain-imputed T1Gd for tumour segmentation | 3D U-Net · BraTS 2023 T1Gd missing | Dice (%): 57.25 -> 61.44 | Improved |
| zero imputation of missing modality instead of synthesis | 3D U-Net · BraTS 2023 mean Dice | Dice (%): 80.06 -> 38.39 | Hurt |
Closest papers by meaning
- Posterior Samplings are Missing Modalities Generators for Medical Image Translation (ECCV 2026 · similarity 0.86)
- OmniBrainBench: A Comprehensive Multimodal Benchmark for Brain Imaging Analysis Across Multi-stage Clinical Tasks (CVPR 2026 · similarity 0.86)
- Null Space Matters: Range-Null Decomposition for Consistent Multi-Contrast MRI Reconstruction (AAAI 2024 · similarity 0.85)
- Enhancing vision-language models for medical imaging: bridging the 3D gap with innovative slice selection (NeurIPS 2024 · similarity 0.85)
- MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI (NeurIPS 2026 · similarity 0.84)
- Score-based Self-supervised MRI Denoising (ICLR 2025 · similarity 0.84)
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.
