Fast library-driven approach for implementation of the voxel spread function technique for correcting magnetic field inhomogeneity artifacts
Purpose: Previously-developed Voxel Spread Function (VSF) method (Yablonskiy, et al, MRM, 2013;70:1283) provides means to correct artifacts induced by macroscopic magnetic field inhomogeneities in the images obtained by multi-Gradient-Recalled-Echo (mGRE) techniques. The goal of this study is to develop a library-driven approach for fast VSF implementation. Methods: The VSF approach describes the contribution of the magnetic field inhomogeneity effects on the mGRE signal decay in terms of the F-function calculated from mGRE phase and magnitude images. A pre-calculated library accounting for a variety of background field gradients caused by magnetic field inhomogeneities was used herein to speed up calculation of the F-function and to generate quantitative R2* maps from the mGRE data collected from two healthy volunteers. Results: As compared with direct calculation of the F-function based on a voxel-wise approach, the new library-driven method substantially reduces computational time from several hours to few minutes, while, at the same time, providing similar accuracy of R2* mapping. Conclusion: The new procedure proposed in this study provides a fast post-processing algorithm that can be incorporated in the quantitative analysis of mGRE data to account for background field inhomogeneity artifacts, thus can facilitate the applications of mGRE-based quantitative techniques in clinical practices.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Pharmacophore-based design by learning on voxel grids
Ligand-based drug discovery (LBDD) relies on making use of known binders to a protein target to find structurally diverse molecules similarly likely to bind. This process typically involves a brute force search of the kn…
Drug DiscoveryJaxUED: A simple and useable UED library in Jax
We present JaxUED, an open-source library providing minimal dependency implementations of modern Unsupervised Environment Design (UED) algorithms in Jax. JaxUED leverages hardware acceleration to obtain on the order of 1…
CPUCatBoost: gradient boosting with categorical features support
In this paper we present CatBoost, a new open-sourced gradient boosting library that successfully handles categorical features and outperforms existing publicly available implementations of gradient boosting in terms of …
ClusteringCPUDimensionality ReductionGeneral Classification+1Baseline: A Library for Rapid Modeling, Experimentation and Development of Deep Learning Algorithms targeting NLP
We introduce Baseline: a library for reproducible deep learning research and fast model development for NLP. The library provides easily extensible abstractions and implementations for data loading, model development, tr…
Deep LearningLanguage ModelingLanguage ModellingMachine Translation+3Multiscale deep context modeling for lossless point cloud geometry compression
We propose a practical deep generative approach for lossless point cloud geometry compression, called MSVoxelDNN, and show that it significantly reduces the rate compared to the MPEG G-PCC codec. Our previous work based …