Radiology: Toolkit compiles anatomy-specific dose data from existing image archive

Researchers have created an open-source informatics toolkit to capture CT radiation dose information from an enterprise image archive, demonstrating that it is possible to create large-scale anatomy-specific radiation exposure data repositories using existing archival information, according to an article published online June 5 in Radiology.

“Before the medical community can engage in systematic and large-scale patient safety initiatives related to radiation exposure, it must first have the tools with which to capture radiation exposure data on a large scale,” wrote Aaron Sodickson, MD, PhD, of Brigham and Women’s Hospital in Boston and colleagues.

The authors noted that dose report screen captures summarizing overall x-ray exposure metrics for a given CT exam already exist in hospital archives. Information about the CT scanner, patient and protocol used is also contained within the DICOM attributes.

“Unfortunately, the data contained in dose screens are not in a format that is inherently accessible to databases, which makes it challenging to locate, extract, validate, and optimize access to the data,” wrote the authors.

To unlock this information, Sodickson and colleagues created the generalized radiation observation kit, or GROK. This toolkit can locate and retrieve volume CT dose index and dose-length product information from a DICOM image archive, as well as convert dose screens to text for analysis.

The authors first validated GROK using 150 randomly selected encounters for each major CT scanner manufacturer, and then turned GROK loose to populate an exposure repository containing nearly 55,000 CT encounters.

Results of the validation process proved GROK’s accuracy. Dose screen retrieval (proportion of correctly located dose screens) occurred at a rate of 99 percent. GROK also correctly extracted and assigned exposure data to specific anatomic regions, such as head, chest or abdomen and pelvis, in 94 percent of cases.

“Large-scale repositories of standardized anatomy-specific exposure data can lead to knowledge discovery by enabling exploration of radiation exposure variations in multiple dimensions, and they can allow monitoring, analysis, and optimization at a regulatory, institutional, scanner, or individual patient level,” wrote the authors.

At the regulatory level, Sodickson et al explained the metrics extracted by the toolkit could help create diagnostic reference levels leading to more robust comparisons between institutions. Within an organization, such information could inform CT protocol quality control efforts and identify outliers or targets for optimization.

This isn’t the first toolkit developed by researchers at Brigham and Women’s Hospital to compile dose information. As previously reported in Health Imaging, researchers demonstrated that exposure data could be extracted from existing nuclear medicine report archives in a study published May 24 in Radiology.