Research map / Image generation / Image restoration
Dynamic PET Image Reconstruction Using Nonnegative Matrix Factorization Incorporated With Deep Image Prior
Tatsuya Yokota, Kazuya Kawai, Muneyuki Sakata, Yuichi Kimura, Hidekata Hontani
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 more parallel U-Nets (1 to 5) in the deep image prior (ablation) | DIP PET reconstruction · Noise-free phantom sinogram | SNR (dB): 28.7 -> 37.3 | Improved |
| use NMF with DIP instead of TV regularization | dynamic PET reconstruction · Brain phantom PET image, 20 dB | SNR (dB): 14.84 -> 18.96 | Improved |
| use NMF with DIP instead of TV regularization | dynamic PET reconstruction · Brain phantom PET image, 10 dB | SNR (dB): 3.07 -> 6.21 | Improved |
| use NMF with DIP instead of EM | dynamic PET reconstruction · Brain phantom PET image, 30 dB | SNR (dB): 3.89 -> 24.62 | Improved |
| use NMF with DIP instead of EM | dynamic PET reconstruction · Brain phantom PET image, noise-free | SNR (dB): 29.43 -> 26.44 | Hurt |
| use NMF with DIP instead of TV | dynamic PET reconstruction · Brain phantom PET image, noise-free | SNR (dB): 22.22 -> 26.44 | Improved |
| use NMF with DIP instead of TV | dynamic PET reconstruction · Brain phantom sinogram, noise-free | SNR (dB): 44.58 -> 40.41 | Hurt |
| use NMF with DIP instead of TV | dynamic PET reconstruction · Brain phantom sinogram, 20 dB | SNR (dB): 29.51 -> 34.76 | Improved |
| use NMF with DIP instead of FBP | dynamic PET reconstruction · Brain phantom PET image, 20 dB | SNR (dB): 0.85 -> 18.96 | Improved |
| run DIP optimization to 20000 iterations instead of stopping at 5000 (ablation) | NMF + DIP · Brain phantom, 20 dB noisy sinogram | reconstruction accuracy | Hurt |
Closest papers by meaning
- A Unified Model for Compressed Sensing MRI Across Undersampling Patterns (CVPR 2025 · similarity 0.80)
- Spatiotemporal-Aware Neural Fields for Dynamic CT Reconstruction (AAAI 2025 · similarity 0.80)
- NPN: Non-Linear Projections of the Null-Space for Imaging Inverse Problems (NeurIPS 2025 · similarity 0.80)
- Deep Rank-One Tensor Functional Factorization for Multi-Dimensional Data Recovery (AAAI 2025 · similarity 0.80)
- UGoDIT: Unsupervised Group Deep Image Prior Via Transferable Weights (NeurIPS 2025 · similarity 0.80)
- Dual Prior Unfolding for Snapshot Compressive Imaging (CVPR 2024 · similarity 0.79)
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.
