Staged and Physics-Grounded Learning Framework with Hyperintensity Prior for Pre-Contrast MRI Synthesis

Contrast-enhanced MRI enhances pathological visualization but often necessitates Pre-Contrast images for accurate quantitative analysis and comparative assessment. However, Pre-Contrast images are frequently unavailable due to time, cost, or safety constraints, or they may suffer from degradation, making alignment challenging. This limitation hinders clinical diagnostics and the performance of tools requiring combined image types. To address this challenge, we propose a novel staged, physics-grounded learning framework with a hyperintensity prior to synthesize Pre-Contrast images directly from Post-Contrast MRIs. The proposed method can generate high-quality Pre-Contrast images, thus, enabling comprehensive diagnostics while reducing the need for additional imaging sessions, costs, and patient risks. To the best of our knowledge, this is the first Pre-Contrast synthesis model capable of generating images that may be interchangeably used with standard-of-care Pre-Contrast images. Extensive evaluations across multiple datasets, sites, anatomies, and downstream tasks demonstrate the model’s robustness and clinical applicability, positioning it as a valuable tool for contrast-enhanced MRI workflows.

MRI scans often use contrast agents to highlight important tissues in the body, helping doctors detect things like tumors or vascular problems. However, these scans usually require a baseline image known as a Pre-Contrast scan for accurate comparison and measurement. Unfortunately, Pre-Contrast scans are sometimes missing, unavailable, or degraded due to time, cost, or patient safety concerns. To solve this issue, we developed a new AI-based method that can recreate high-quality Pre-Contrast MRI images using only the contrast-enhanced scan. The method is grounded in image physics and uses a staged learning process to remove the effects of the contrast enhancement from the image. This approach can reduce the need for repeat scans, saving time, cost, and patient discomfort. It also helps improve the accuracy of tools that rely on both image types. We tested our method across many different body parts, diseases, and scanners, and found it to be accurate, reliable, and fast, making it a valuable addition to real-world medical imaging workflows.