MRIReco.jl: An MRI reconstruction framework written in Julia

  title={MRIReco.jl: An MRI reconstruction framework written in Julia},
  author={Tobias Knopp and Mirco Grosser},
  journal={Magnetic Resonance in Medicine},
  pages={1633 - 1646}
  • T. Knopp, M. Grosser
  • Published 29 January 2021
  • Physics, Computer Science, Medicine
  • Magnetic Resonance in Medicine
The aim of this work is to develop a high‐performance, flexible, and easy‐to‐use MRI reconstruction framework using the scientific programming language Julia. 
2 Citations
Advances in spiral fMRI: A high-resolution study with single-shot acquisition
It is demonstrated that high-resolution spiral fMRI at 7 T is not only feasible, but delivers both excellent image quality, BOLD sensitivity, and spatial specificity of the activation maps, with little artifactual blurring.
Vendor-neutral sequences (VENUS) and fully transparent workflows improve inter-vendor reproducibility of quantitative MRI
It is concluded that vendor-neutral workflows are feasible and compatible with clinical MRI scanners and the significant reduction of inter-vendor variability using VENUS has important implications for qMRI research and for the reliability of multicenter clinical trials.


Gadgetron: An open source framework for medical image reconstruction
This work presents a new open source framework for medical image reconstruction called the “Gadgetron.” The framework implements a flexible system for creating streaming data processing pipelines
Accelerating advanced mri reconstructions on gpus
The acceleration afforded by the GPU greatly increases the appeal of the advanced reconstruction for clinical MRI applications and dramatically reduces the reconstruction's required bandwidth to on-chip memory.
Real-Time Reconstruction of Sensitivity Encoded Radial Magnetic Resonance Imaging Using a Graphics Processing Unit
A reconstruction algorithm for commodity graphics hardware (GPUs) is presented to enable real time reconstruction of sensitivity encoded radial imaging (radial SENSE) and it is demonstrated that a radial profile order based on the golden ratio facilitates reconstruction from an arbitrary number of profiles.
SENSE: Sensitivity encoding for fast MRI
The problem of image reconstruction from sensitivity encoded data is formulated in a general fashion and solved for arbitrary coil configurations and k‐space sampling patterns and special attention is given to the currently most practical case, namely, sampling a common Cartesian grid with reduced density.
A Note on the Iterative MRI Reconstruction from Nonuniform k-Space Data
A recently proposed implicit discretisation scheme is used which generalises the standard approach based on gridding and improves the reconstruction quality when the sampling scheme or the weights are less regular.
Sampling density compensation in MRI: Rationale and an iterative numerical solution
This method uses only the coordinates of the sampled data; unlike previous methods, it does not require knowledge of the trajectories and can easily handle trajectories that “cross” in k‐space.
Field Inhomogeneity Correction Based on Gridding Reconstruction for Magnetic Resonance Imaging
Three new algorithms, compatible with a direct conjugate phase and an iterative algebraic reconstruction, are derived from a straightforward embedding of the data into a higher dimensional space and one of them promises to provide a favorable compromise between fidelity and complexity compared with existing algorithms.
Advances in sensitivity encoding with arbitrary k‐space trajectories
Using the proposed method, SENSE becomes practical with nonstandard k‐space trajectories, enabling considerable scan time reduction with respect to mere gradient encoding, and the in vivo feasibility of non‐Cartesian SENSE imaging with iterative reconstruction is demonstrated.
Human-sized magnetic particle imaging for brain applications
A human-sized MPI device with low technical requirements designed for detection of brain ischemia is presented, which opens up a variety of medical applications and would allow monitoring of stroke on intensive care units.
PyTorch: An Imperative Style, High-Performance Deep Learning Library
This paper details the principles that drove the implementation of PyTorch and how they are reflected in its architecture, and explains how the careful and pragmatic implementation of the key components of its runtime enables them to work together to achieve compelling performance.