Learning Scan-Adaptive MRI Undersampling Patterns with Pre-Optimized Mask Supervision
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914049466302464 |
|---|---|
| author | Dhar, Aryan Gautam, Siddhant Ravishankar, Saiprasad |
| author_facet | Dhar, Aryan Gautam, Siddhant Ravishankar, Saiprasad |
| contents | Deep learning techniques have gained considerable attention for their ability to accelerate MRI data acquisition while maintaining scan quality. In this work, we present a convolutional neural network (CNN) based framework for learning undersampling patterns directly from multi-coil MRI data. Unlike prior approaches that rely on in-training mask optimization, our method is trained with precomputed scan-adaptive optimized masks as supervised labels, enabling efficient and robust scan-specific sampling. The training procedure alternates between optimizing a reconstructor and a data-driven sampling network, which generates scan-specific sampling patterns from observed low-frequency $k$-space data. Experiments on the fastMRI multi-coil knee dataset demonstrate significant improvements in sampling efficiency and image reconstruction quality, providing a robust framework for enhancing MRI acquisition through deep learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_16846 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning Scan-Adaptive MRI Undersampling Patterns with Pre-Optimized Mask Supervision Dhar, Aryan Gautam, Siddhant Ravishankar, Saiprasad Image and Video Processing Deep learning techniques have gained considerable attention for their ability to accelerate MRI data acquisition while maintaining scan quality. In this work, we present a convolutional neural network (CNN) based framework for learning undersampling patterns directly from multi-coil MRI data. Unlike prior approaches that rely on in-training mask optimization, our method is trained with precomputed scan-adaptive optimized masks as supervised labels, enabling efficient and robust scan-specific sampling. The training procedure alternates between optimizing a reconstructor and a data-driven sampling network, which generates scan-specific sampling patterns from observed low-frequency $k$-space data. Experiments on the fastMRI multi-coil knee dataset demonstrate significant improvements in sampling efficiency and image reconstruction quality, providing a robust framework for enhancing MRI acquisition through deep learning. |
| title | Learning Scan-Adaptive MRI Undersampling Patterns with Pre-Optimized Mask Supervision |
| topic | Image and Video Processing |
| url | https://arxiv.org/abs/2509.16846 |