In a review of volume CT, Dr. Frank John Parrish from MIA Victoria in Malvern, Australia, noted that the first CT scanner, which appeared in the 1970s, could take one slice every two seconds. Now, volume-CT scanners can take 192 slices per second, and even faster scanners are undergoing clinical evaluation.

Multislice volume scanners have reduced the relatively high radiation doses associated with thin slices. Although thin slices require higher radiation doses than thick slices, several studies have shown that they can reveal more disease characteristics. Still, on a case-by-case basis, radiologists must consider whether the greater diagnostic information warrants the higher radiation dose associated with thin slices.

Current CT scanners can generate more slices per rotation and provide radiologists with a “bewildering array” of postprocessing options and potentially thousands of images to review, Parrish stated. The sheer number has made the use of film impossible and thus has increased the use of digital technology. Images, especially thin slices, also can occupy hundreds of gigabytes of disk space, a problem that can be solved by off-loading them onto servers designed for handling them.

Several ways to display or to reconstruct CT images can improve diagnostic performance. In the past, only MRI could display images in any plane, but now CT can do the same, an ability for which there is no downside. More accurate diagnoses can be enabled by a technique called volume rendering and by the relatively new field of computer-aided diagnosis, as well as by computer packages such as those for virtual endoscopy, for brain perfusion imaging and for blood vessel analysis. A 3-D reconstruction technique called maximum intensity projection can make biological structures such as blood vessels more visible, but some evidence suggests that it also can obscure structures. (American Journal of Roentgenology, September 2007, pp. 528-534.)