A call for structure

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
Lisa Fratt, Editor

Inconsistency pervades stroke imaging protocols, while the lack of integration prevents content-based image retrieval (CBIR) from more widespread implementation.

Relatively unrelated at first glance, this month’s portal exclusives hint at a larger issue. That is, in many cases imaging remains all too subjective.

CBIR offers a potential means to harness visual image data, and the model could prove promising for CAD by automatically identifying and preliminarily organizing relevant prior studies for comparison and analysis.

One hitch, however, is the lack of integration between CBIR and PACS workflow. Most CBIR systems are standalone, which may represent an untended repeat of the implementation path of advanced visualization technology. Specifically, the standalone workstation model impedes workflow, and thus, discourages and decelerates more widespread adoption until the technology matures into a more open, integrated system.

German researchers have developed a prototype CBIR viewer that leverages DICOM standard 118 to overcome early integration challenges. Developers believe the viewer could bridge the gap between radiologists’ workflow requirements and CBIR.

Meanwhile, researchers investigating stroke protocols have concluded that physicians and researchers employ inconsistent CT- and MR-based perfusion values in identification of tissue at risk of infarction in acute stroke.

“The target region of tissue for reperfusion therapy in ischemic stroke is that which is hypoperfused and at risk of infarction, but still viable–the “ischemic penumbra”?as salvage of this tissue may improve functional outcome. This tissue needs to be distinguished from definitely nonviable and from not at-risk/normal tissue,” explained Krishna A. Dani, MBChB, MRCP, from the University of Glasgow Southern General Hospital in Scotland, and colleagues, in a study published online in June in Annals of Neurology.

The researchers found multiple different definitions of viable and at-risk tissue for both CT- and MR-based protocols, leading them to suggest that physicians not make clinical decisions about stroke treatment based on threshold values at the present time.

They did suggest that researchers leverage existing perfusion imaging data to develop agreed-upon standards.

Although imaging has made tremendous strides in the last two decades, it still needs tighter integration and more uniform standards. What gaps do you see? And how are you filling those gaps?

Please let us know.

Lisa Fratt, editor