Huang Haoyuan, Liu Baoer, Xu Yikai, Zhou Wu
School of Medical Information Engineering, Guangzhou University of Chinese Medicine, Guangzhou, China.
Department of Medical Imaging Center, Nanfang Hospital, Southern Medical University, Guangzhou, China.
Med Phys. 2023 Mar;50(3):1614-1622. doi: 10.1002/mp.16031. Epub 2023 Jan 14.
Intravoxel incoherent motion (IVIM) is a type of diffusion-weighted imaging (DWI), and IVIM model parameters (water molecule diffusion rate D , pseudo-diffusion coefficient D , and tissue perfusion fraction F ) have been widely used in the diagnosis and characterization of malignant lesions.
This study proposes a deep-learning model with synthetic-to-real domain adaptation to fit the IVIM model parameters of DWI.
Ninety-eight consecutive patients diagnosed with hepatocellular carcinoma between January 2017 and September 2020 were included in the study, and routine IVIM-DWI serial examinations were performed using a 3.0 T magnetic resonance imaging system in preoperative MR imaging. The proposed method is mainly composed of two modules: a convolutional neural network-based IVIM model fitting network to map b-value images to the IVIM parameter maps and a domain discriminator to improve the accuracy of the IVIM parameter maps in the real data. The proposed method was compared with previously reported fitting methods, including the nonlinear least squares (NLSs), IVIM-NET , and self-supervised U-network methods. The IVIM parameter-fitting performance was assessed by measuring the DWI reconstruction performance and testing the robustness of each method against noise using noise-corrupted data.
The DWI reconstruction performance demonstrates that the proposed method has better reconstruction accuracy for DWI with a low signal-to-noise ratio, which implies that the proposed method improves the fitting accuracy of the IVIM parameters. Noise-corrupt experiments show that the proposed method is more robust against noise-corrupted signals. With the proposed method, no outliers were found in D , and outliers were reduced for F in the abnormal regions (proposed method: 1.85%; NLS: 5.90%; IVIM-NET : 6.61%; and self-U-net: 25.36%). Moreover, experiments show that the proposed method has a more stable parameter estimation performance than the existing methods in the absence of real data.
IVIM parameters can be estimated using a synthetic-to-real domain-adaptation framework with deep learning, and the proposed method outperforms previously reported methods.
体素内不相干运动（IVIM）是一种扩散加权成像（DWI），IVIM模型参数（水分子扩散率D、伪扩散系数D*和组织灌注分数F）已广泛应用于恶性病变的诊断和特征描述。
本研究提出一种具有合成到真实域适应的深度学习模型，以拟合DWI的IVIM模型参数。
本研究纳入了2017年1月至2020年9月期间连续诊断为肝细胞癌的98例患者，并在术前磁共振成像中使用3.0T磁共振成像系统进行常规IVIM-DWI序列检查。所提出的方法主要由两个模块组成：一个基于卷积神经网络的IVIM模型拟合网络，用于将b值图像映射到IVIM参数图；一个域判别器，用于提高真实数据中IVIM参数图的准确性。将所提出的方法与先前报道的拟合方法进行比较，包括非线性最小二乘法（NLSs）、IVIM-NET和自监督U网络方法。通过测量DWI重建性能并使用噪声污染数据测试每种方法对噪声的鲁棒性来评估IVIM参数拟合性能。
DWI重建性能表明，所提出的方法对低信噪比的DWI具有更好的重建精度，这意味着所提出的方法提高了IVIM参数的拟合精度。噪声污染实验表明，所提出的方法对噪声污染信号更具鲁棒性。使用所提出的方法，在D中未发现异常值，在异常区域中F的异常值有所减少（所提出的方法：1.85%；NLS：5.90%；IVIM-NET：6.61%；自监督U网络：25.36%）。此外，实验表明，在所没有真实数据的情况下，所提出的方法比现有方法具有更稳定的参数估计性能。
可以使用具有深度学习的合成到真实域适应框架来估计IVIM参数，并且所提出的方法优于先前报道的方法。
