2021/01/05 by Zaccharie Ramzi, Jean‐Luc Starck, Ramzi, Zaccharie +3 · 1 citation
Engineering · Medicine · #Advanced MRI Techniques and Applications #Advanced X-ray and CT Imaging #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #FOS: Physical sciences #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Medical Imaging Techniques and Applications #Medical Physics (physics.med-ph) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2101.01570
openalex publication_date 2021/01/05 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Deep neural networks have recently been thoroughly investigated as a powerful\ntool for MRI reconstruction. There is a lack of research, however, regarding\ntheir use for a specific setting of MRI, namely non-Cartesian acquisitions. In\nthis work, we introduce a novel kind of deep neural networks to tackle this\nproblem, namely density compensated unrolled neural networks, which rely on\nDensity Compensation to correct the uneven weighting of the k-space. We assess\ntheir efficiency on the publicly available fastMRI dataset, and perform a small\nablation study. Our results show that the density-compensated unrolled neural\nnetworks outperform the different baselines, and that all parts of the design\nare needed. We also open source our code, in particular a Non-Uniform Fast\nFourier transform for TensorFlow.\n