Contrast-enhanced MRI synthesis using dense-dilated residual convolutions based 3D network toward elimination of gadolinium in neuro-oncology.

Abstract

Recent studies have raised broad safety and health concerns about using of gadolinium contrast agents during magnetic resonance imaging (MRI) to enhance identification of active tumors. In this paper, we developed a deep learning-based method for three-dimensional (3D) contrast-enhanced T1-weighted (T1) image synthesis from contrast-free image(s). The MR images of 1251 patients with glioma from the RSNA-ASNR-MICCAI BraTS Challenge 2021 dataset were used in this study. A 3D dense-dilated residual U-Net (DD-Res U-Net) was developed for contrast-enhanced T1 image synthesis from contrast-free image(s). The model was trained on a randomly split training set (n=800) using a customized loss function and validated on a validation set (n=200) to improve its generalizability. The generated images were quantitatively assessed against the ground-truth on a test set (n=251) using the mean absolute error (MAE), mean-squared error (MSE), peak signal-to-noise ratio (PSNR), structural similarity (SSIM), normalized mutual information (NMI), and Hausdorff distance (HDD) metrics. We also performed a qualitative visual similarity assessment between the synthetic and ground-truth images. The effectiveness of the proposed model was compared with a 3D U-Net baseline model and existing deep learning-based methods in the literature. Our proposed DD-Res U-Net model achieved promising performance for contrast-enhanced T1 synthesis in both quantitative metrics and perceptual evaluation on the test set (n=251). Analysis of results on the whole brain region showed a PSNR (in dB) of 29.882±5.924, a SSIM of 0.901±0.071, a MAE of 0.018±0.013, a MSE of 0.002±0.002, a HDD of 2.329±9.623, and a NMI of 1.352±0.091 when using only T1 as input; and a PSNR (in dB) of 30.284±4.934, a SSIM of 0.915±0.063, a MAE of 0.017±0.013, a MSE of 0.001±0.002, a HDD of 1.323±3.551, and a NMI of 1.364±0.089 when combining T1 with other MRI sequences. Compared to the U-Net baseline model, our model revealed superior performance. Our model demonstrated excellent capability in generating synthetic contrast-enhanced T1 images from contrast-free MR image(s) of the whole brain region when using multiple contrast-free images as input. Without incorporating tumor mask information during network training, its performance was inferior in the tumor regions compared to the whole brain which requires further improvements to replace the gadolinium administration in neuro-oncology.