Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.25348 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914598981992448 |
|---|---|
| author | Radhakrishnan, Veera Varuni Dinesh, Chinthaka Azim, Qurat-ul-Ain |
| author_facet | Radhakrishnan, Veera Varuni Dinesh, Chinthaka Azim, Qurat-ul-Ain |
| contents | Low-dose computed tomography (LDCT) reconstruction faces a critical tradeoff between reconstruction quality and resource requirements. While recent deep learning methods achieve state-of-the-art performance, they typically rely on over 500,000 parameters trained on large-scale datasets exceeding 35,000 scans. This work investigates whether graph-based regularization can provide meaningful noise reduction under strict resource constraints. We propose Deep Graph Laplacian Regularization (Deep GLR), integrating quadratic graph regularization into a Proximal Forward-Backward Splitting optimization framework with three lightweight CNN modules. Evaluated on the LoDoPaB-CT benchmark, Deep GLR achieves 30.70 dB PSNR, representing a 6.33 dB improvement over filtered backprojection, while using only 91,848 parameters trained on 1000 samples (2.8\% of standard training set). Compared to benchmark methods, this represents 5.8 times better parameter efficiency and 30 times better data efficiency per dB improvement. The learned graph bandwidth parameter ($ε$=1.25) converges to interpretable values, suggesting the method captures meaningful image priors rather than overfitting. While a 13 dB gap remains versus state-of-the-art methods, results demonstrate that graph-based regularization provides a favorable efficiency-quality tradeoff for resource-constrained medical imaging scenarios. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_25348 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Parameter-Efficient CT Reconstruction via Deep Graph Laplacian Regularization Radhakrishnan, Veera Varuni Dinesh, Chinthaka Azim, Qurat-ul-Ain Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning Symbolic Computation Low-dose computed tomography (LDCT) reconstruction faces a critical tradeoff between reconstruction quality and resource requirements. While recent deep learning methods achieve state-of-the-art performance, they typically rely on over 500,000 parameters trained on large-scale datasets exceeding 35,000 scans. This work investigates whether graph-based regularization can provide meaningful noise reduction under strict resource constraints. We propose Deep Graph Laplacian Regularization (Deep GLR), integrating quadratic graph regularization into a Proximal Forward-Backward Splitting optimization framework with three lightweight CNN modules. Evaluated on the LoDoPaB-CT benchmark, Deep GLR achieves 30.70 dB PSNR, representing a 6.33 dB improvement over filtered backprojection, while using only 91,848 parameters trained on 1000 samples (2.8\% of standard training set). Compared to benchmark methods, this represents 5.8 times better parameter efficiency and 30 times better data efficiency per dB improvement. The learned graph bandwidth parameter ($ε$=1.25) converges to interpretable values, suggesting the method captures meaningful image priors rather than overfitting. While a 13 dB gap remains versus state-of-the-art methods, results demonstrate that graph-based regularization provides a favorable efficiency-quality tradeoff for resource-constrained medical imaging scenarios. |
| title | Parameter-Efficient CT Reconstruction via Deep Graph Laplacian Regularization |
| topic | Image and Video Processing Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning Symbolic Computation |
| url | https://arxiv.org/abs/2605.25348 |