Eidex Zach, Safari Mojtaba, Qiu Richard L J, Yu David S, Shu Hui-Kuo, Mao Hui, Yang Xiaofeng
Department of Radiation Oncology, Emory University, Atlanta, Georgia, USA.
Winship Cancer Institute, Emory University, Atlanta, Georgia, USA.
Med Phys. 2025 Apr;52(4):2064-2073. doi: 10.1002/mp.17600. Epub 2024 Dec 23.
Gadolinium-based contrast agents (GBCAs) are commonly used in MRI scans of patients with gliomas to enhance brain tumor characterization using T1-weighted (T1W) MRI. However, there is growing concern about GBCA toxicity. This study develops a deep-learning framework to generate T1-postcontrast (T1C) from pre-contrast multiparametric MRI.
We propose the tumor-aware vision transformer (TA-ViT) model that predicts high-quality T1C images. The predicted tumor region is significantly improved (p < 0.001) by conditioning the transformer layers from predicted segmentation maps through the adaptive layer norm zero mechanism. The predicted segmentation maps were generated with the multi-parametric residual (MPR) ViT model and transformed into a latent space to produce compressed, feature-rich representations. The TA-ViT model was applied to T1w and T2-FLAIR to predict T1C MRI images of 501 glioma cases from an open-source dataset. Selected patients were split into training (N = 400), validation (N = 50), and test (N = 51) sets. Model performance was evaluated with the peak-signal-to-noise ratio (PSNR), normalized cross-correlation (NCC), and normalized mean squared error (NMSE).
Both qualitative and quantitative results demonstrate that the TA-ViT model performs superior against the benchmark MPR-ViT model. Our method produces synthetic T1C MRI with high soft tissue contrast and more accurately synthesizes both the tumor and whole brain volumes. The synthesized T1C images achieved remarkable improvements in both tumor and healthy tissue regions compared to the MPR-ViT model. For healthy tissue and tumor regions, the results were as follows: NMSE: 8.53 ± 4.61E-4; PSNR: 31.2 ± 2.2; NCC: 0.908 ± 0.041 and NMSE: 1.22 ± 1.27E-4, PSNR: 41.3 ± 4.7, and NCC: 0.879 ± 0.042, respectively.
The proposed method generates synthetic T1C images that closely resemble real T1C images. Future development and application of this approach may enable contrast-agent-free MRI for brain tumor patients, eliminating the risk of GBCA toxicity and simplifying the MRI scan protocol.
钆基造影剂（GBCAs）常用于胶质瘤患者的MRI扫描，以通过T1加权（T1W）MRI增强脑肿瘤特征。然而，人们对GBCA毒性的担忧日益增加。本研究开发了一种深度学习框架，用于从造影前的多参数MRI生成T1加权造影后（T1C）图像。
我们提出了肿瘤感知视觉Transformer（TA-ViT）模型，该模型可预测高质量的T1C图像。通过自适应层归一化零机制根据预测的分割图对Transformer层进行调节，预测的肿瘤区域得到了显著改善（p < 0.001）。预测的分割图由多参数残差（MPR）ViT模型生成，并转换到潜在空间以产生压缩的、特征丰富的表示。TA-ViT模型应用于T1W和T2-FLAIR，以预测来自开源数据集的501例胶质瘤病例的T1C MRI图像。选定的患者被分为训练集（N = 400）、验证集（N = 50）和测试集（N = 51）。使用峰值信噪比（PSNR）、归一化互相关（NCC）和归一化均方误差（NMSE）评估模型性能。
定性和定量结果均表明，TA-ViT模型的性能优于基准MPR-ViT模型。我们的方法生成的合成T1C MRI具有高软组织对比度，并且能更准确地合成肿瘤和全脑体积。与MPR-ViT模型相比，合成的T1C图像在肿瘤和健康组织区域均有显著改善。对于健康组织和肿瘤区域，结果如下：NMSE：8.53 ± 4.61E-4；PSNR：31.2 ± 2.2；NCC：0.908 ± 0.041，以及NMSE：1.22 ± 1.27E-4，PSNR：41.3 ± 4.7，NCC：0.879 ± 0.042。
所提出方法生成的合成T1C图像与真实T1C图像非常相似。该方法的未来发展和应用可能使脑肿瘤患者无需使用造影剂进行MRI检查，消除GBCA毒性风险并简化MRI扫描方案。
