Enhancing quality and speed in database-free neural network reconstructions of undersampled MRI with SCAMPI

Abstract

Purpose: We present SCAMPI (Sparsity Constrained Application of deep Magnetic resonance Priors for Image reconstruction), an untrained deep Neural Network for MRI reconstruction without previous training on datasets. It expands the Deep Image Prior approach with a multidomain, sparsity-enforcing loss function to achieve higher image quality at a faster convergence speed than previously reported methods. Methods: Two-dimensional MRI data from the FastMRI dataset with Cartesian undersampling in phase-encoding direction were reconstructed for different acceleration rates for single coil and multicoil data. Results: The performance of our architecture was compared to state-of-the-art Compressed Sensing methods and ConvDecoder, another untrained Neural Network for two-dimensional MRI reconstruction. SCAMPI outperforms these by better reducing undersampling artifacts and yielding lower error metrics in multicoil imaging. In comparison to ConvDecoder, the U-Net architecture combined with an elaborated loss-function allows for much faster convergence at higher image quality. SCAMPI can reconstruct multicoil data without explicit knowledge of coil sensitivity profiles. Moreover, it is a novel tool for reconstructing undersampled single coil k-space data. Conclusion: Our approach avoids overfitting to dataset features, that can occur in Neural Networks trained on databases, because the network parameters are tuned only on the reconstruction data. It allows better results and faster reconstruction than the baseline untrained Neural Network approach.

arXiv comment14 pages, 6 figures, 4 tables. Supplementary materials with 4 figures, 1 table

References in corpus (12)

• An overview of deep learning in medical imaging focusing on MRI
• On instabilities of deep learning in image reconstruction - Does AI come at a cost?
• Adaptive Diffusion Priors for Accelerated MRI Reconstruction
• Computed Tomography Reconstruction Using Deep Image Prior and Learned Reconstruction Methods
• RARE: Image Reconstruction using Deep Priors Learned without Ground Truth
• Unsupervised Deep Learning Methods for Biological Image Reconstruction and Enhancement
• ENLIVE: An Efficient Nonlinear Method for Calibrationless and Robust Parallel Imaging
• ENSURE: A General Approach for Unsupervised Training of Deep Image Reconstruction Algorithms
• Image Restoration using Total Variation Regularized Deep Image Prior
• An untrained deep learning method for reconstructing dynamic magnetic resonance images from accelerated model-based data
• Non-Learning based Deep Parallel MRI Reconstruction (NLDpMRI)
• Federated Learning of Generative Image Priors for MRI Reconstruction