Region‐guided focal adversarial learning for CT‐to‐MRI translation: A proof‐of‐concept and validation study in hepatocellular carcinoma
Yi‐Fan Xia, Meng Zeng, Shuwen Sun, Qiuping Liu, Jiulou Zhang, Rui Zhi, Fei‐Yu Lu, Wei Chen, Yu‐Dong Zhang · Medical Physics · 2025
BACKGROUND: Generative adversarial networks (GANs) have recently demonstrated significant potential for producing virtual images with the same characteristics as real-life landscapes, thereby enhancing various medical tasks. PURPOSE: To design a region-guided focal GAN (Focal-GAN) for translating images between CT and MRI and test its clinical applicability in patients with hepatocellular carcinoma (HCC). METHODS: Between January 2012 and October 2021, two cohorts of patients with HCC who underwent contrast-enhanced CT (Center 1, n = 685) and MRI (Center 1, n = 516; Center 2, n = 318) were retrospectively enrolled. We trained the Focal-GAN model by adding tumor regions to a baseline Cycle-GAN framework to steer the model toward focal attention learning. The quality of the images generated was assessed using an open-source MRQy tool. The clinical applicability of the Focal-GAN was evaluated by applying the nnUNet and ResNet-50 model for tumor segmentation and microvascular invasion (MVI) prediction in HCC on the generated images. RESULTS: In the ablation tests, Focal-GAN achieved a higher fidelity than the conventional Cycle-GAN in the generated image quality assessment with MRQy. Regarding applicability, regardless of tumor size, nnUNet trained with focal-GAN-generated images achieved higher Dice scores than nnUNet trained using Cycle-GAN-generated images for HCC segmentation in both internal (0.607 vs. 0.341, p < 0.01) and external (0.796 vs. 0.753, p < 0.001) validation. Additionally, ResNet-50 trained with Focal-GAN-generated images produced higher areas-under-curve (AUCs) than ResNet-50 trained with real images for MVI prediction in both internal (0.754 vs. 0.665, p = 0.048) and external (0.670 vs. 0.579, p < 0.001) validation. CONCLUSIONS: The designed Focal-GAN model can generate virtual MR images from unpaired CT images, thereby extending the clinical applicability of CT in the liver tumor diagnostic pathway.