Research map / 3D vision / Face and shape
QSMnet-INR: Single-Orientation Quantitative Susceptibility Mapping via Implicit Neural Representation in k-Space
Xuan Cai, Ruomi Guo, Kangyuan Zheng, Jiaqi Wang, Xiao‐Wen Luo, Jing Zhao, Hongbin Han, Mengting Liu
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 INR dipole kernel completion module to QSMnet (ablation) | QSMnet · 2016 QSM Challenge | PSNR (dB): 40.999 -> 43.821 | Improved |
| add INR dipole kernel completion module to QSMnet (ablation) | QSMnet · 2016 QSM Challenge | HFEN: 0.591 -> 0.403 | Improved |
| add INR dipole kernel completion module to QSMnet (ablation) | QSMnet · 2016 QSM Challenge | SSIM: 0.898 -> 0.946 | Improved |
| add cone-null dipole loss to QSMnet + INR (ablation) | QSMnet + INR · 2016 QSM Challenge | PSNR (dB): 43.821 -> 45.669 | Improved |
| add cone-null dipole loss to QSMnet + INR (ablation) | QSMnet + INR · 2016 QSM Challenge | NRMSE: 0.644 -> 0.521 | Improved |
| add cone-null dipole loss to QSMnet + INR (ablation) | QSMnet + INR · 2016 QSM Challenge | SSIM: 0.946 -> 0.967 | Improved |
| raise gradient loss weight from 0.1 to 0.2 | QSMnet-INR · 2016 QSM Challenge | PSNR (dB): 45.67 -> 44.90 | Hurt |
| raise dipole loss weight from 0.3 to 0.4 | QSMnet-INR · 2016 QSM Challenge | PSNR (dB): 45.67 -> 44.80 | Hurt |
| lower dipole loss weight from 0.3 to 0.2 (with wmodel 0.5) | QSMnet-INR · 2016 QSM Challenge | PSNR (dB): 45.67 -> 45.90 | Mixed |
| raise model weight from 0.4 to 0.6 | QSMnet-INR · 2016 QSM Challenge | PSNR (dB): 45.67 -> 45.39 | Mixed |
| QSMnet-INR versus DIAM-CNN | DIAM-CNN · Multi-Orientation GRE (single orientation) | PSNR (dB): 39.734 -> 40.252 | Improved |
| QSMnet-INR versus QSMnet | QSMnet · 2016 QSM Challenge | HFEN: 0.591 -> 0.362 | Improved |
Closest papers by meaning
- Equivariant spatio-hemispherical networks for diffusion MRI deconvolution (NeurIPS 2024 · similarity 0.83)
- QUTCC: Quantile Uncertainty Training and Conformal Calibration for Imaging Inverse Problems (NeurIPS 2026 · similarity 0.83)
- Adaptive Sampling of k-Space in Magnetic Resonance for Rapid Pathology Prediction (ICML 2024 · similarity 0.82)
- Boosting ViT-based MRI Reconstruction from the Perspectives of Frequency Modulation, Spatial Purification, and Scale Diversification (AAAI 2025 · similarity 0.82)
- Null Space Matters: Range-Null Decomposition for Consistent Multi-Contrast MRI Reconstruction (AAAI 2024 · similarity 0.82)
- From Pixels to Views: Learning Angular-Aware and Physics-Consistent Representations for Light Field Microscopy (NeurIPS 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.
