New techniques soup up MRI

Magnetic resonance imaging is a powerful diagnostic tool that lets doctors peer deep inside the human body. However, its pace can be slow, with each image taking many seconds or minutes to build.

Now a team of researchers at the University of Wisconsin–Madison, led by medical physicist Charles Mistretta, has patented a suite of technologies that promise to soup up MRI, making it speedy enough to catch fleeting events during procedures like angiography, while creating exceptionally vivid, three-dimensional images with less background interference.

These more nimble MRI techniques will mean fewer uncomfortable minutes inside scanners for patients and much quicker access to high-quality diagnostic images for physicians, Mistretta says.

The information used to build MRI images exists in an alternate data universe called k-space; conventional MRI methodically assembles each image from the entire set of k-space data. By sampling only critical elements of the k-space data and combining the sampled parts in ingenious ways, Mistretta and his colleagues can construct images much faster, without skimping on quality.

Faster image generation makes MRI applicable to angiography – the visualization of arteries and veins – where the best image captures the brief instant when a contrast-enhancing chemical begins to course through a blood vessel. Like a photographer who shoots rapid-fire pictures during a fast-paced basketball game to get the one perfect photo for the sports page, Mistretta’s technique grabs a series of images in quick succession, ensuring that one will provide an ideal diagnostic view.

With the help of his technology, MRI should completely replace more risky X-ray approaches to angiography, Mistretta says. And many other flight plans for navigating k-space exist, providing virtually endless possibilities for applications, such as motion-correcting algorithms for times when patients squirm in the scanner.