Categories: world

Artificial intelligence can help reduce gadolinium dose in MR

CHICAGO, November 26, 2018 / PRNewswire / – Researchers use artificial intelligence to reduce the dose of a contrast agent…

CHICAGO, November 26, 2018 / PRNewswire / – Researchers use artificial intelligence to reduce the dose of a contrast agent that can be left in the body after MRI exams, according to a study presented today at the annual meeting of the Radiological Society in North America ( RSNA).

Gadolinium is a heavy metal used in contrast media that enhances images on MR. Recent studies have shown that the trace amounts of the metal remain in the bodies of people who have undergone some types of gadolinium. The effects of this disposal are not known, but radiologists are working proactively to optimize patient safety while retaining the important information that Gadolinium Enhanced MRI Scanning provides.

Story Continues Below Advertisement

Ad Statistics

Times viewed: 971

Number of visits: 44

Questions to ask and consider about MRI reel processes

What to expect from MR coil Repairs that meet clinical expectations and driving returns. Click for a guideline for standards for expecting and demanding sustainable MR repair.

“There is concrete evidence that gadolinium deposits in the brain and the body,” said study leader for researcher Enhao Gong, Ph.D., researcher at Stanford University in Stanford, Calif. “Although the consequences of this are unclear, it is important to mitigate potential patient risks while the maximum clinical value of the MRI exam is crucial.”

Dr. Gong and colleagues in Stanford have studied deep learning as a way to achieve this goal. Deep learning is a sophisticated artificial intelligence technique that teaches computers with examples. By using models called drop-down networks, the computer not only recognizes images but also finds subtle differences between image data that a human observer may not be able to detect.

To train the deep learning algorithm, researchers used MR images from 200 patients who had received contrast-enhanced MRI tests for different indications. The combined three sets of images for each patient: before contrast scans, performed before counter-administration and called zero-dose scans; low dose scans, acquired after 10 percent of the standard administration of gadolinium dose; and fulldos scans, acquired after 100% administration of the dose.

The algorithm learned to approximate full-length scans from nolldos and low-dose images. Neuroradiologists then evaluated the images for contrast enhancement and overall quality.

The result showed that the image quality was not significantly different between the low-dose, algorithm-enhanced MRI images and the fully-dot, contrast-enhanced MRI images. The initial results also showed the potential for creating the equivalent of full-dose, contrast-enhanced MRI images without the use of contrast agents.

Published by