Fast chemical exchange saturation transfer imaging based on PROPELLER acquisition and deep neural network reconstruction

Abstract

Purpose To develop a method for fast chemical exchange saturation transfer (CEST) imaging. Methods The periodically rotated overlapping parallel lines enhanced reconstruction (PROPELLER) sampling scheme was introduced to shorten the acquisition time. Deep neural network was employed to reconstruct CEST contrast images. Numerical simulation and experiments on a creatine phantom, hen egg, and in vivo tumor rat brain were performed to test the feasibility of this method. Results The results from numerical simulation and experiments show that there is no significant difference between reference images and CEST‐PROPELLER reconstructed images under an acceleration factor of 8. Conclusion Although the deep neural network is trained entirely on synthesized data, it works well on reconstructing experimental data. The proof of concept study demonstrates that the combination of the PROPELLER sampling scheme and the deep neural network enables considerable acceleration of saturated image acquisition and may find applications in CEST MRI.

Whole‐brain amide CEST imaging at 3T with a steady‐state radial MRI acquisition

“…Recently, a radial sampling CEST method combining PROPELLER readout was proposed on rat brain. 28 However, due to the use of a conventional labeling scheme, that is, continuous-wave labeling followed by PROPELLER readout, this method was unable to reduce the total scan time.…”

“…To further improve the steady‐state saturation with radial readout chemical exchange saturation transfer (starCEST) SNR, multilinear singular value decomposition (MLSVD) 27 was applied. Recently, a radial sampling CEST method combining PROPELLER readout was proposed on rat brain 28 . However, due to the use of a conventional labeling scheme, that is, continuous‐wave labeling followed by PROPELLER readout, this method was unable to reduce the total scan time.…”

Deep Learning and Its Application to Function Approximation for MR in Medicine: An Overview

“…In the field of chemical exchange saturation transfer (CEST), DNNs approximated a function that received images with several frequency offsets as the inputs, and produced frequency offset images at 3 ppm and -3 ppm for generating a CEST image as the outputs; 128 and a function that received several images with different frequency offsets and under-sampling patterns as the inputs, and produced a CEST image as the output. 129 In the latter method, 129 input images were acquired using periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER). 130…”

Super-resolved Reconstruction Method for Spatiotemporally Encoded MRI Based on Deep Neural Network

“…Imaging, SPEN MRI)是一种新兴的超快速成像技术 [1][2][3] 样"机制提供了可用于 SR 重建的冗余信息 [4] ．基于此，近十年来，已有数种针 对于单扫描 SPEN MRI 的 SR 重建方法被提出，如共轭梯度下降法 [4] 、部分傅里 叶法 [5] 、去卷积法 [6] 和超分辨率增强边缘鬼影去除法(SEED) [7] 据快速完成重建 [9,[15][16][17]19] ．2015 年，Ronneberger 等人提出一种新型 DNN 结构 U-Net [20]…”

“…建上表现优异 [21,22] ． 在此背景下， 本文提出了一种基于 U-Net 的单扫描 SPEN MRI SR 重建方法．该方法根据实际 SPEN MRI 实验条件制作大量模拟训练样本，采 用模拟样本对网络进行训练 [9,17] ，然后通过训练完成的网络模型对预处理后的…”