Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany.
Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany.
NMR Biomed. 2021 Apr;34(4):e4474. doi: 10.1002/nbm.4474. Epub 2021 Jan 21.
Quantitative Na magnetic resonance imaging (MRI) provides tissue sodium concentration (TSC), which is connected to cell viability and vitality. Long acquisition times are one of the most challenging aspects for its clinical establishment. K-space undersampling is an approach for acquisition time reduction, but generates noise and artifacts. The use of convolutional neural networks (CNNs) is increasing in medical imaging and they are a useful tool for MRI postprocessing. The aim of this study is Na MRI acquisition time reduction by k-space undersampling. CNNs were applied to reduce the resulting noise and artifacts. A retrospective analysis from a prospective study was conducted including image datasets from 46 patients (aged 72 ± 13 years; 25 women, 21 men) with ischemic stroke; the Na MRI acquisition time was 10 min. The reconstructions were performed with full dataset (FI) and with a simulated dataset an image that was acquired in 2.5 min (RI). Eight different CNNs with either U-Net-based or ResNet-based architectures were implemented with RI as input and FI as label, using batch normalization and the number of filters as varying parameters. Training was performed with 9500 samples and testing included 400 samples. CNN outputs were evaluated based on signal-to-noise ratio (SNR) and structural similarity (SSIM). After quantification, TSC error was calculated. The image quality was subjectively rated by three neuroradiologists. Statistical significance was evaluated by Student's t-test. The average SNR was 21.72 ± 2.75 (FI) and 10.16 ± 0.96 (RI). U-Nets increased the SNR of RI to 43.99 and therefore performed better than ResNet. SSIM of RI to FI was improved by three CNNs to 0.91 ± 0.03. CNNs reduced TSC error by up to 15%. The subjective rating of CNN-generated images showed significantly better results than the subjective image rating of RI. The acquisition time of Na MRI can be reduced by 75% due to postprocessing with a CNN on highly undersampled data.
定量钠磁共振成像（MRI）提供组织钠浓度（TSC），与细胞活力和活力有关。采集时间长是其临床应用最具挑战性的方面之一。K 空间欠采样是一种减少采集时间的方法，但会产生噪声和伪影。卷积神经网络（CNN）在医学成像中的应用越来越多，它们是 MRI 后处理的有用工具。本研究的目的是通过 K 空间欠采样减少 Na MRI 的采集时间。将 CNN 应用于减少由此产生的噪声和伪影。对一项前瞻性研究的回顾性分析包括来自 46 名（年龄 72±13 岁；25 名女性，21 名男性）缺血性脑卒中患者的图像数据集； Na MRI 的采集时间为 10 分钟。重建分别采用全数据集（FI）和模拟的 2.5 分钟采集数据集（RI）。使用 RI 作为输入，FI 作为标签，实现了 8 种不同的基于 U-Net 或 ResNet 的 CNN，均使用批量归一化和滤波器数量作为变化参数。使用 9500 个样本进行训练，使用 400 个样本进行测试。根据信噪比（SNR）和结构相似性（SSIM）评估 CNN 输出。量化后，计算 TSC 误差。三位神经放射学家对图像质量进行主观评分。通过学生 t 检验评估统计学意义。平均 SNR 为 21.72±2.75（FI）和 10.16±0.96（RI）。U-Net 将 RI 的 SNR 提高到 43.99，因此性能优于 ResNet。三个 CNN 将 RI 与 FI 的 SSIM 提高到 0.91±0.03。CNN 将 TSC 误差降低了高达 15%。CNN 生成的图像的主观评分明显优于 RI 的主观图像评分。通过对高度欠采样数据进行 CNN 后处理，可将 Na MRI 的采集时间减少 75%。
