Motivation: Contrast-enhanced MR images with gadolinium-based contrast agents is crucial for brain tumor evaluation, but contraindicated in some patients. AI-based synthetic contrast-enhanced MR images offer an alternative, though studies have mainly focused on gliomas. Goal(s): To generate synthetic contrast-enhanced MR images from noncontrast MR images across various brain tumor types using multi-center datasets. Approach: A 3D-UNet-based AI model was developed using noncontrast T1-weighted and T2-weighted images as input and contrast-enhanced T1-weighted images as target. Results: Synthetic contrast-enhanced MR images demonstrated high SSIM and PSNR across various brain tumor types, with significant correlation of tumor contrast-enhanced volume compared to real ones. Impact: Our study suggests potential clinical applicability of AI-based synthetic contrast-enhanced MR images generated from noncontrast MR images across various brain tumor types, offering a promising alternative for patients unable to receive gadolinium-based contrast agents.
Takita et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: