Tracking low dose performance: How IR impacts low-contrast resolution

Reduction of contrast dosage by 25 percent or more on CT scans results in reduced low-contrast spatial resolution and loss of detection of 6mm nodules on accreditation phantoms using iterative reconstruction (IR) techniques, according to results published online March 26 in the journal Radiology.

Advancements in IR have made the technology an attractive alternative to traditional full-dose filtered back projection on CT due to its ability to substantially reduce noise and maintain high-contrast spatial resolution. The same cannot be said for IR when used to image low-contrast objects such as liver lesions or certain regions of the brain, according to lead author Cynthia McCollough, PhD, and her colleagues at the Mayo Clinic in Rochester, Minn. “It is critical to determine how much low-contrast spatial resolution is affected by IR, such that as dose is reduced, the noise reduction caused by IR does not compromise the ability to detect and characterize low-contrast objects,” wrote McCollough et al. “Thus, the dose reduction potential of IR is highly dependent on the diagnostic task.”

The Mayo Clinic researchers set out to determine just how much of a reduction in contrast dosage could be achieved without degrading low-contrast spatial resolution performance. To do so, they performed scans on the American College of Radiology (ACR) CT accreditation phantom LCR section using two scanner models at volume CT dose indexes of 8, 12, and 16 mGy.  Both filtered back projection and two manufacturers’ IR techniques were used each at “moderate” and “strong” strengths. The images were reviewed and graded by three diagnostic medical physicists to assess the various techniques’ abilities to image 6mm nodules in the ACR phantom.

Their results showed that  low-contrast spatial resolution was reduced with every reduction in contrast dosage, with reductions of 25-50 percent producing images that were inferior to traditional flitered back projection. “The improved appearance of the images in terms of decreased noise levels, can conceal this fact, as can quantitative measurements, such as contrast-to-noise ratios,” McCullough et al wrote. “For radiation dose reductions … the ability to visualize all four 6-mm low-contrast rods in the American College of Radiology CT accreditation phantom can be lost.”

McCullough and her team believe their findings are of high clinical importance in helping doctors better determine when dose reductions can be successfully achieved. “It is imperative that the ability to detect and characterize low-contrast lesions not be compromised in the pursuit of reduced radiation dose levels,” the authors wrote. “Until sufficient studies have been performed to show that clinically relevant diagnostic performance for low-contrast imaging tasks is maintained or improved at reduced dose levels, use of IR techniques to decrease the appearance of image noise may increase the likelihood of missing clinically important findings, particularly those that are subtle in appearance.”