Department of Radiology and Nuclear Medicine, Cancer Center Amsterdam, Amsterdam UMC, University of Amsterdam, HV Amsterdam 1081, the Netherlands.
Centre for Big Data Research in Health, UNSW, Sydney, Australia.
Med Image Anal. 2022 Aug;80:102512. doi: 10.1016/j.media.2022.102512. Epub 2022 Jun 7.
Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is an MRI technique for quantifying perfusion that can be used in clinical applications for classification of tumours and other types of diseases. Conventionally, the non-linear least squares (NLLS) methods is used for tracer-kinetic modelling of DCE data. However, despite promising results, NLLS suffers from long processing times (minutes-hours) and noisy parameter maps due to the non-convexity of the cost function. In this work, we investigated physics-informed deep neural networks for estimating physiological parameters from DCE-MRI signal-curves. Three voxel-wise temporal frameworks (FCN, LSTM, GRU) and two spatio-temporal frameworks (CNN, U-Net) were investigated. The accuracy and precision of parameter estimation by the temporal frameworks were evaluated in simulations. All networks showed higher precision than the NLLS. Specifically, the GRU showed to decrease the random error on v by a factor of 4.8 with respect to the NLLS for noise (SD) of 1/20. The accuracy was better for the prediction of the v parameter in all networks compared to the NLLS. The GRU and LSTM worked with arbitrary acquisition lengths. The GRU was selected for in vivo evaluation and compared to the spatio-temporal frameworks in 28 patients with pancreatic cancer. All neural network approaches showed less noisy parameter maps than the NLLS. The GRU had better test-retest repeatability than the NLLS for all three parameters and was able to detect one additional patient with significant changes in DCE parameters post chemo-radiotherapy. Although the U-Net and CNN had even better test-retest characteristics than the GRU, and were able to detect even more responders, they also showed potential systematic errors in the parameter maps. Therefore, we advise using our GRU framework for analysing DCE data.
动态对比增强磁共振成像（DCE-MRI）是一种定量灌注的 MRI 技术，可用于肿瘤分类和其他类型疾病的临床应用。传统上，使用非线性最小二乘法（NLLS）方法对 DCE 数据进行示踪动力学建模。然而，尽管取得了有希望的结果，但 NLLS 由于代价函数的非凸性，存在处理时间长（分钟-小时）和参数图噪声的问题。在这项工作中，我们研究了基于物理的深度神经网络，以从 DCE-MRI 信号曲线估计生理参数。研究了三种体素-wise 时间框架（FCN、LSTM、GRU）和两种时空框架（CNN、U-Net）。在模拟中评估了时间框架参数估计的准确性和精度。所有网络的参数估计精度均高于 NLLS。具体来说，GRU 相对于 NLLS，在噪声（SD）为 1/20 的情况下，将 v 的随机误差降低了 4.8 倍。与 NLLS 相比，所有网络对 v 参数的预测准确性都更好。GRU 和 LSTM 可以处理任意采集长度。选择 GRU 进行体内评估，并与 28 名胰腺癌患者的时空框架进行比较。与 NLLS 相比，所有神经网络方法的参数图噪声都更小。与 NLLS 相比，GRU 对所有三个参数的测试-重测重复性更好，并且能够检测到一名在化疗-放疗后 DCE 参数有明显变化的额外患者。虽然 U-Net 和 CNN 的测试-重测特征甚至比 GRU 更好，并且能够检测到更多的应答者，但它们在参数图中也显示出潜在的系统误差。因此，我们建议使用我们的 GRU 框架来分析 DCE 数据。
