Medical Image Processing Research Group (MIPRG), Electrical & Computer Engineering Department, COMSATS University Islamabad, Islamabad, Pakistan.
Service of Radiology, Faculty of Medicine, Geneva University Hospitals, University of Geneva, Geneva, Switzerland.
J Digit Imaging. 2023 Feb;36(1):276-288. doi: 10.1007/s10278-022-00709-5. Epub 2022 Nov 4.
Under-sampling in diffusion-weighted imaging (DWI) decreases the scan time that helps to reduce off-resonance effects, geometric distortions, and susceptibility artifacts; however, it leads to under-sampling artifacts. In this paper, diffusion-weighted MR image (DWI-MR) reconstruction using deep learning (DWI U-Net) is proposed to recover artifact-free DW images from variable density highly under-sampled k-space data. Additionally, different optimizers, i.e., RMSProp, Adam, Adagrad, and Adadelta, have been investigated to choose the best optimizers for DWI U-Net. The reconstruction results are compared with the conventional Compressed Sensing (CS) reconstruction. The quality of the recovered images is assessed using mean artifact power (AP), mean root mean square error (RMSE), mean structural similarity index measure (SSIM), and mean apparent diffusion coefficient (ADC). The proposed method provides up to 61.1%, 60.0%, 30.4%, and 28.7% improvements in the mean AP value of the reconstructed images in our experiments with different optimizers, i.e., RMSProp, Adam, Adagrad, and Adadelta, respectively, as compared to the conventional CS at an acceleration factor of 6 (i.e., AF = 6). The results of DWI U-Net with the RMSProp, Adam, Adagrad, and Adadelta optimizers show 13.6%, 10.0%, 8.7%, and 8.74% improvements, respectively, in terms of mean SSIM with respect to the conventional CS at AF = 6. Also, the proposed technique shows 51.4%, 29.5%, 24.04%, and 18.0% improvements in terms of mean RMSE using the RMSProp, Adam, Adagrad, and Adadelta optimizers, respectively, with reference to the conventional CS at AF = 6. The results confirm that DWI U-Net performs better than the conventional CS reconstruction. Also, when comparing the different optimizers in DWI U-Net, RMSProp provides better results than the other optimizers.
在扩散加权成像（DWI）中进行欠采样可以减少扫描时间，有助于降低离共振效应、几何变形和磁化率伪影；然而，它会导致欠采样伪影。在本文中，提出了一种使用深度学习（DWI U-Net）对扩散加权磁共振图像（DWI-MR）进行重建的方法，以便从变量密度的高度欠采样 k 空间数据中恢复无伪影的 DW 图像。此外，研究了不同的优化器，即 RMSProp、Adam、Adagrad 和 Adadelta，以选择用于 DWI U-Net 的最佳优化器。将重建结果与传统的压缩感知（CS）重建进行了比较。使用平均伪影功率（AP）、均方根误差（RMSE）、结构相似性指数测量（SSIM）和表观扩散系数（ADC）来评估恢复图像的质量。与传统 CS 相比，在不同的优化器（即 RMSProp、Adam、Adagrad 和 Adadelta）下，所提出的方法在我们的实验中提供了高达 61.1%、60.0%、30.4%和 28.7%的重建图像平均 AP 值的改进，在加速因子为 6（即 AF=6）时。在使用 RMSProp、Adam、Adagrad 和 Adadelta 优化器的 DWI U-Net 中，与传统 CS 相比，在 AF=6 时，平均 SSIM 分别提高了 13.6%、10.0%、8.7%和 8.74%。此外，与传统 CS 相比，在 AF=6 时，使用 RMSProp、Adam、Adagrad 和 Adadelta 优化器的 DWI U-Net 分别在平均 RMSE 方面提高了 51.4%、29.5%、24.04%和 18.0%。结果证实，DWI U-Net 的性能优于传统 CS 重建。此外，在 DWI U-Net 中比较不同的优化器时，RMSProp 提供的结果优于其他优化器。
