Researchers Rev Up Low-Dose CT

An unfortunate byproduct of CT scans is the possibility of exposing patients to high doses of radiation, which can increase cancer risk. For that reason, many experts recommend low-dose CT scanners. Unfortunately, low-dose CT scans require a half an hour or more of subsequent image processing time. A new $1.9-million study at the University of Michigan seeks to combat the problem with a new low-dose CT scanning system that can process the CT scans in just five minutes.

In a recent news release from the University of Michigan, details were revealed about the new study led by Jeffrey Fessler, a professor of electrical and computer engineering at the University of Michigan. Fessler spoke about the challenges of exposing patients to radiation through CT scanning, and the importance of reducing the exposure.

"It's known that a radiation dose can increase the risk of cancer, but nobody knows exactly how much," Fessler said.  

Unfortunately we do have an idea of how much radiation a CT scan can impose upon patients. On average, a CT scan can expose patients to a radiation dose comparable to about five to eight months' worth of radiation one might naturally receive in their daily life. For this reason, the National Institute of Health has prioritized the reduction of radiation doses associated with these scans, which served as the catalyst for Fessler's study.

His team began by investigating methods that could reduce the amount of radiation exposure by more than half the dosage, a considerable amount considering today's modern technology. The drawback, however, is that these new low-dosage scans produce images that are not nearly as crisp as those produced by some of the higher-dosage machines.

"Reducing the radiation is like setting a very short shutter speed on your camera," Fessler said. "You'd get a grainy picture, but you could use Photoshop to try to get a better image. The data processing needed for low-dose x-ray CT is far more complicated."

Increasing processing times and improving scanning images have risen to the forefront of cancer research, as researchers continue to search for innovative ways to improve cancer detection methods. Often times improving one of these aforementioned aspects can come at the expense of the other. In the case of CT scans, faster processing usually results in lower resolution images.

These days, it can take anywhere from 30 minutes to an hour to reconstruct the images for diagnosis in a CT scan, which is frustratingly impractical given that the actual scan only takes a few minutes. From the outset, Fessler and his team aimed to cut the processing time down to just five minutes, without compromising the image resolution too much.

The team's solution was to work with multicore computing, that can take advantage of advanced algorithms that can divide the data among the various processors, allowing each to handle a specific region, before stitching the image back together to produce a final image.

If the study proves successful, the work could lead to low-dose CT scans becoming a quick and easy screening option for patients, while limiting their exposure to potentially dangerous radiation levels. It will also serve as crucial for patients who need multiple screenings to track the progress of tumors within the body.

"In repeat scan situations, it's crucial that the dose be very low," Fessler said. With Fessler's study looking to cut the current dosage in half, these new low-dosage CT scans could be the wave of the future.

Kristopher Sturgis is a contributor to Qmed and MPMN.

Like what you're reading? Subscribe to our daily e-newsletter.