Roger Luechinger | Searcharxiv

arXiv2025

A unipolar head gradient for high-field MRI without encoding ambiguity

Purpose: MRI gradients with a conventional, bipolar design generally face a trade-off between performance, encoding ambiguity, and circumventing the latter by means of RF selectivity. This problem is particularly limiting in cutting-edge brain imaging performed at field strengths >= 7T and using high-performance head gradients. Methods: To address this issue, the present work proposes to fundamentally eliminate the encoding ambiguity in head gradients by using a unipolar z-gradient design that takes advantage of the signal-free range on one side of the imaging volume. This concept is demonstrated by implementation of a unipolar head gradient for operation at 7T. Results: Imaging in phantoms and in vivo demonstrates elimination of backfolding due to encoding ambiguity. At the same time, the unipolar design achieves efficiency on par with conventional bipolar design, resulting in high amplitude and slew-rate performance. Conclusion: The prospect of gradient systems based on a unipolar design holds promise for all advanced neuroimaging that demands high gradient performance. It will make the greatest difference at 7T and beyond where the absence of ambiguity removes a key concern and constraint in terms of RF behaviour and instrumentation.

physics.med-ph↗

arXiv2018

MR image reconstruction using deep density priors

Algorithms for Magnetic Resonance (MR) image reconstruction from undersampled measurements exploit prior information to compensate for missing k-space data. Deep learning (DL) provides a powerful framework for extracting such information from existing image datasets, through learning, and then using it for reconstruction. Leveraging this, recent methods employed DL to learn mappings from undersampled to fully sampled images using paired datasets, including undersampled and corresponding fully sampled images, integrating prior knowledge implicitly. In this article, we propose an alternative approach that learns the probability distribution of fully sampled MR images using unsupervised DL, specifically Variational Autoencoders (VAE), and use this as an explicit prior term in reconstruction, completely decoupling the encoding operation from the prior. The resulting reconstruction algorithm enjoys a powerful image prior to compensate for missing k-space data without requiring paired datasets for training nor being prone to associated sensitivities, such as deviations in undersampling patterns used in training and test time or coil settings. We evaluated the proposed method with T1 weighted images from a publicly available dataset, multi-coil complex images acquired from healthy volunteers (N=8) and images with white matter lesions. The proposed algorithm, using the VAE prior, produced visually high quality reconstructions and achieved low RMSE values, outperforming most of the alternative methods on the same dataset. On multi-coil complex data, the algorithm yielded accurate magnitude and phase reconstruction results. In the experiments on images with white matter lesions, the method faithfully reconstructed the lesions. Keywords: Reconstruction, MRI, prior probability, machine learning, deep learning, unsupervised learning, density estimation

cs.CV↗