A Wisconsin start-up company has filed for a patent for a technique that uses deep-learning technology to produce contrast-free MR images. The firm believes the technique could someday eliminate the use of gadolinium-based contrast agents.
IQ-AI of Milwaukee filed for the patent for a gadolinium-free MRI technique that uses as its input noncontrast MR images, collected with sequences such as T1-weighted, T2-weighted, fluid-attenuated inverse-recovery (FLAIR), and/or diffusion-weighted imaging (DWI) protocols.
The algorithm uses these inputs to create what the company calls simulated T1 + contrast (T1+C) images, without the use of an actual contrast agent. The image would be created within a few seconds and would eliminate the operational challenges involved in contrast administration, such as gadolinium deposition, allergic reactions, and workflow issues.
