Research map / 3D vision / Human pose estimation
Reliable Evaluation of MRI Motion Correction: Dataset and Insights
Kun Wang, Tobit Klug, Stefan Ruschke, Jan Kirschke, Reinhard Heckel
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 |
|---|---|---|---|
| use feature-space metric MoMRISim instead of pixel metrics for reference-based evaluation | PMoC3D paired real-world evaluation · PMoC3D moderate/severe motion | Spearman correlation with PMAS: MoMRISim reaches 0.95, highest correlation | Improved |
| pixel-wise metrics (PSNR, SSIM, AP) for evaluating motion correction | PMoC3D paired evaluation · PMoC3D | correlation with PMAS | No clear effect |
| use reference-based evaluation when motion is mild | PMoC3D paired evaluation · PMoC3D mild motion | reliability of reference | Hurt |
| evaluate on simulated motion instead of real motion | MotionTTT · severe motion, real vs simulated | residual artifacts: real worse in 75% of comparisons, similar in 25% | Hurt |
| use reference-free VLM score (GPT-4o) instead of gradient metrics TG and AES | AltOpt and MotionTTT reconstructions · PMoC3D | alignment with PMAS | Improved |
| use reference-free VLM score for ranking including E2E stacked U-net | VLM score · PMoC3D | correlation with PMAS | Hurt |
| use DISTS for reference-based evaluation | stacked U-net reconstructions · PMoC3D | correlation with PMAS | Hurt |
| run MotionTTT phase 2 splitting corrupted shots in addition to phase 1 (ablation) | MotionTTT · PMoC3D | reconstruction quality: saves 1-5% of read-out lines | No clear effect |
Closest papers by meaning
- MotionTTT: 2D Test-Time-Training Motion Estimation for 3D Motion Corrected MRI (NeurIPS 2024 · similarity 0.87)
- Moner: Motion Correction in Undersampled Radial MRI with Unsupervised Neural Representation (ICLR 2025 · similarity 0.86)
- MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI (NeurIPS 2026 · similarity 0.83)
- Motion Artifact Removal in Pixel-Frequency Domain via Alternate Masks and Diffusion Model (AAAI 2025 · similarity 0.82)
- Autoregressive Image Diffusion: Generation of Image Sequence and Application in MRI (NeurIPS 2024 · similarity 0.82)
- Efficient Noise Calculation in Deep Learning-based MRI Reconstructions (ICML 2025 · similarity 0.82)
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.
