Mike Bassett, Associate Editor

While radiologists are ready and willing to take advantage of the benefits provided by advances in medical technology, without adequate training and an understanding of what that technology can do, it loses much of its utility.

A study published in this month's Radiology discusses one of the issues associated with the use of computer-aided detection (CAD) software during CT colonography (CTC)—that in fact, many times the sensitivity of stand-alone CAD is greater than that of CAD plus a reader.

“[R]adiologists sometimes incorrectly dismiss lesions annotated appropriately by CAD,” wrote Stuart A. Taylor, MD, and his colleagues from University College Hospital in London. The question they wanted to answer is why radiologists engage in what appears to be “apparently illogical behavior.”

Unsurprisingly, the authors determined that the use of CTC without CAD enabled readers to more easily identify large polyps (10 mm in diameter or higher), those that were rated easy to visualize, and those that were uncoated with by tagged fluid.

But what was really interesting--and, as the authors pointed out, counterintuitive—is that the odds of correctly identifying polyps by using CAD actually decreased as the size of the polyp increased. “[T]he larger the polyp, the more likely it was to be misclassified by readers as a false-positive finding,” they found.

Why? They suggested several reasons. For example, they said their data showed that most radiologists are able to detect large lesions without CAD and might leave polyps that are more difficult to characterize, such as those that are flat or have an irregular shape.

They authors wrote that it might be useful to have CAD output modified to additionally flag polyps that are larger or are coated by tagged fluid so that it will guarantee reader scrutiny. The benefits of that approach will have to await further study.

The point is, the authors wrote, is that while it may be useful to have CAD become more ”intelligent” to increase effectiveness, the underlying principle still stands-- using CAD can be beneficial. But, they wrote, those using it must not only be trained to use it but also to distinguish between true- and false-positive CAD prompts. If the technology is going to be used, it's critical that the person using it is competent.

If you or your group is interested in finding out more about the capabilities of advanced visualization technology, and how it can extend and expand the reach of diagnostic imaging service lines, head over to our Healthcare TechGuide and check out the variety of systems offered there.

Lastly, if you have a comment or report to share about how the utilization of advanced visualization technology is changing your practice, please contact me at the address below. I look forward to hearing from you

Michael Bassett, Associate Editor
mbassett@trimedmedia.com