Taking CAD to the next level

More than one decade since CAD first made its way into the breast imaging landscape, the software continues to evolve. Two studies highlighted in this month’s advanced visualization portal demonstrate its power and potential. One suggests it can improve CT colonography reading efficiency, while another points the way toward better breast imaging algorithms.

Despite many radiologists’ interest in CT colonography, the exam does not yet have widespread adoption or reimbursement. Barriers include including high interobserver variability, a time-consuming image review process and lack of cost-effectiveness.

But a double-reading model that employs CAD as an initial read, followed by radiologist review, provides detection capabilities on par with primary review by a radiologist, while also boosting efficiency, reported Gabriella Iussich, MD, from the Institute for Cancer Research and Treatment in Turin, Italy, and colleagues April 29 in Radiology. Not a bad deal.

Meanwhile, mammography remains the problematic gold standard in breast cancer detection, partially because some tumors are exceedingly difficult to detect. Until recently, little data about the characteristics of these troublesome tumors existed.

Mohammad A. Rawashdeh, MSc, from the University of Sydney in Australia, and colleagues shed a bit of light on the situation via a study devised to determine if features such as mass size, shape, texture, contrast and density reduced detection.

“The results here demonstrate that breast density, lesion size, and lesion shape are the key features that affect detection of breast masses,” the researchers wrote in the May issue of Academic Radiology.

They also observed that masses with irregular margins or greater levels of spiculation have a lower detectability index, which prompted the suggestion that these findings are integrated into the development of new CAD algorithms.

How might these findings impact your practice? Please email and let us know.

Finally, for the just plain cool, be sure to check out two stories at the end of the newsletter: “Video: MR makes perfect fit for amputees,” and “3D printer used to replace 75 percent of man’s skull.”

Lisa Fratt, editor