Department of Diagnostic and Interventional Radiology, Eberhard Karls University Tuebingen, Hoppe-Seyler-Strasse 3, 72076 Tuebingen, Germany.
MR Application Predevelopment, Siemens Healthcare GmbH, Erlangen, Germany.
Acad Radiol. 2024 Mar;31(3):921-928. doi: 10.1016/j.acra.2023.06.035. Epub 2023 Jul 25.
To determine the impact on acquisition time reduction and image quality of a deep learning (DL) reconstruction for accelerated diffusion-weighted imaging (DWI) of the pelvis at 1.5 T compared to standard DWI.
A total of 55 patients (mean age, 61 ± 13 years; range, 27-89; 20 men, 35 women) were consecutively included in this retrospective, monocentric study between February and November 2022. Inclusion criteria were (1) standard DWI (DWI) in clinically indicated magnetic resonance imaging (MRI) at 1.5 T and (2) DL-reconstructed DWI (DWI). All patients were examined using the institution's standard MRI protocol according to their diagnosis including DWI with two different b-values (0 and 800 s/mm) and calculation of apparent diffusion coefficient (ADC) maps. Image quality was qualitatively assessed by four radiologists using a visual 5-point Likert scale (5 = best) for the following criteria: overall image quality, noise level, extent of artifacts, sharpness, and diagnostic confidence. The qualitative scores for DWI and DWI were compared with the Wilcoxon signed-rank test.
The overall image quality was evaluated to be significantly superior in DWI compared to DWI for b = 0 s/mm, b = 800 s/mm, and ADC maps by all readers (P < .05). The extent of noise was evaluated to be significantly less in DWI compared to DWI for b = 0 s/mm, b = 800 s/mm, and ADC maps by all readers (P < .001). No significant differences were found regarding artifacts, lesion detectability, sharpness of organs, and diagnostic confidence (P > .05). Acquisition time for DWI was 2:06 minutes, and simulated acquisition time for DWI was 1:12 minutes.
DL image reconstruction improves image quality, and simulation results suggest that a reduction in acquisition time for diffusion-weighted MRI of the pelvis at 1.5 T is possible.
在 1.5T 磁共振成像（MRI）中，与标准弥散加权成像（DWI）相比，深度学习（DL）重建对加速骨盆弥散加权成像（DWI）采集时间的缩短和图像质量的影响。
本研究为回顾性、单中心研究，共纳入 2022 年 2 月至 11 月期间 55 例连续患者（平均年龄 61±13 岁，范围 27-89 岁，男 20 例，女 35 例）。纳入标准为：（1）在临床指征的 1.5T MRI 中进行标准 DWI（DWI）；（2）DL 重建的 DWI（DWI）。所有患者均根据诊断使用机构标准 MRI 协议进行检查，包括两种不同 b 值（0 和 800 s/mm）的 DWI 和表观扩散系数（ADC）图的计算。4 名放射科医生使用视觉 5 分李克特量表（5 分为最佳）对以下标准进行定性评估：整体图像质量、噪声水平、伪影程度、清晰度和诊断信心。DWI 和 DWI 的定性评分采用 Wilcoxon 符号秩检验进行比较。
所有读者均认为 DWI 的整体图像质量在 b=0 s/mm、b=800 s/mm 和 ADC 图上均优于 DWI（P<0.05）。所有读者均认为 DWI 的噪声程度在 b=0 s/mm、b=800 s/mm 和 ADC 图上均显著低于 DWI（P<0.001）。在伪影、病变检出率、器官清晰度和诊断信心方面无显著差异（P>0.05）。DWI 的采集时间为 2:06 分钟，DWI 的模拟采集时间为 1:12 分钟。
DL 图像重建可提高图像质量，模拟结果表明，1.5T 磁共振成像中骨盆弥散加权成像的采集时间可能会缩短。
