NCI projects aim to lower the barriers to shared cancer-related imaging data

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

Two major projects being highlighted this week at RSNA from the National Cancer Institute (NCI) look to widen access to cancer-related imaging information done for clinical trials so that radiologists and others can easily share the data in a free and open source manner. The projects, called NCI caBIG Imaging Workspace and the National Cancer Imaging Archive, are being represented at RSNA this week in Chicago with 22 talks and 12 exhibits in all, along with an assortment of other useful resource. Look for the display at the Lakeside Learning Center.

“It’s a really fascinating set of exhibits and talks because for the first time that I know of in a major organized concerted effort there is funding from the government through NCI for advancing medical informatics to give the ability to exchange information of all different types from an informatics approach,” said Eliot L. Siegel, MD, professor and vice-chair of Radiology; University of Maryland School of Medicine, Department of Radiology, and chief of imaging for VA Maryland Healthcare System; and head of imaging informatics for the National Cancer Institute. “There is a huge community of people in academia, vendors, and a number of other ones all cooperating and participating in this neutral ground that NCI has set up to advance imaging informatics. Also the display really focuses on what’s happening specifically for clinical trials for research at the National Cancer Institute, a lot of the things we’re doing have implications for really all of imaging. And so I think it’s going to be a really interesting eye opener for a lot of people that come,” Siegel added.

The caBIG Imaging Workspace has been in development for nearly a year and has involved a wide assortment of hundreds of people throughout the industry in an inclusive, very collaborative fashion “to solve imaging informatics challenges on all fronts,” Siegel said.

Several developments will be on display all week at RSNA and visitors will be able to discuss and interact and learn more about the projects.

All the things that we do are completely free and completely open source so that anybody in the medical imaging community – vendors, people in academia, people in research, clinical folks – are all encouraged to use all of the software or all of the things that we’re developing,” said Siegel. “So we’re really anxious to get feedback from people in the imaging community, particularly those based outside of the United States, to get an understanding of what their needs and requirements are.”

Key demonstrations:

X IP – eXtensible Imaging Platform: A free and open source set of software and an imaging platform that will give users a set of tools as part of an image interpretation workstation to do reading associated with research or clinical trials. The platform “will be the equivalent of a PACS image interpretation workstation for researchers who are doing image interpretation that will be free and open source,” said Siegel. Moreover, it is a “platform with standards so that people who end up developing algorithms for image processing or image enhancement or visualization or quantitative measurement can develop them so that people from around the world will be able to use it,” Siegel added.

In the current situation, every vendor uses a different algorithm which comes up with a different result. Siegel believes that visitors will be excited to learn about the potential of the open source platform for broad applications, from research all the way to potential clinical use at some point.

X IP is being developed in conjunction with Siemens Medical Solutions. In fact, each of the projects on display has been awarded to an academic medical center who in many cases is working with a partner such as Siemens.

AIM – Annotation and Image Mark-up: AIM is an attempt to create a standard or agreement as to how images are labeled or annotated, and how information on an image can be connected with knowledge in the outside world, Siegel said. He gave the example of circling of a lung nodule. “The fact that it has been circled means I’m saying that the pixels contained in that circle represent something important. If I label it as a lung nodule and if I measure it I’m providing knowledge or information about that region or portion of the images. That image can then be annotated or marked-up in many different ways by different people” through the Workspace, said Siegel. On the other hand, most proprietary workstations from