In an article slated to be published in the journal Magnetic Resonance in Medicine, researchers, led by Dr. Elfar Adalsteinsson, an associate professor of electrical engineering and computer science and health sciences and technology, and Dr. Vivek Goyal, an MIT associate professor of electrical engineering and computer science, from the Massachusetts Institute of Technology (MIT; Cambridge, MA, USA) described an algorithm they have developed to greatly speed up the MRI process. The algorithm uses information gained from the first contrast scan to help it generate the subsequent images. In this way, the scanner does not have to start from the beginning each time it produces a different image from the raw data, but already has a basic outline to work from, considerably shortening the time it takes to acquire each later scan.

To devise this outline, the software looks for features that are common to all the different scans, such as the basic anatomic structure. Specifically, the algorithm utilizes the first scan to predict the likely position of the boundaries between different types of tissue in the subsequent contrast scans. “Given the data from one contrast, it gives you a certain likelihood that a particular edge, say the periphery of the brain or the edges that confine different compartments inside the brain, will be in the same place,” Dr. Adalsteinsson stated.

However, the algorithm cannot impose too much information from the first scan onto the subsequent ones, Dr. Goyal noted, as this would risk losing the unique tissue features revealed by the different contrasts. “You don’t want to presuppose too much,” he said. “So you don’t assume, for example, that the bright-and-dark pattern from one image will be replicated in the next image, because in fact those kinds of dark and light patterns are often reversed, and can reveal completely different tissue properties.”

So for each pixel, the algorithm calculates what new information it needs to construct the image, and what data--such as the edges of different types of tissue--it can take from the earlier scans, according to graduate student and first author Berkin Bilgic. The result is an MRI scan that is three times faster to complete, slashing the time patients spend in the scanner from 45 to 15 minutes. This faster scan time does have a slight impact on image quality, Mr. Bilgic admitted, but it is much better than competing algorithms.

The scientists are now working to improve the algorithm by speeding up the time it takes to process the raw image data into a final scan that can be assessed by clinicians, once the patient has stepped out of the MRI scanner. Using standard computer processors, this final step currently takes considerably longer than with traditional MRI scans. However, the researchers believe they can reduce this calculation time down to the same as that of traditional MRI scans using recent developments in computing hardware from the gaming industry. “Graphics processing units, or GPUs, are orders of magnitude faster at certain computational tasks than general processors, like the particular computational task that we need for this algorithm,” Dr. Adalsteinsson said. A student at the laboratory is now working to implement the algorithm on a dedicated GPU, he reported.

Dwight Nishimura, the director of the magnetic resonance systems research laboratory at Stanford University (Stanford, CA, USA), stated that Dr. Adalsteinsson’s group has conducted very noteworthy algorithmic research. “This work is potentially of high significance because it applies to routine clinical MRI, among other applications,” he remarked. “Ultimately, their approach might enable a substantial reduction in examination time.”

Related Links:
Massachusetts Institute of Technology