Purpose or Case Report: Functional magnetic resonance urography (fMRU) provides not only anatomic information but quantitative functional information as well. Several free tools are available to perform functional analysis. One such tool, CHOP-fMRU, is appealing because its creators have provided extensive instructions on their software and scan protocols.

Several barriers were encountered in attempting to implement CHOP-fMRU at OHSU. First, to perform the analysis, the appropriate DICOM files must be acquired from PACS. CHOP-fMRU is compatible with a limited number of transfer syntaxes, so a conversion is required after PACS download if the encoding is incompatible. Second, CHOP-fMRU is incompatible with some types of data elements that can be found in files with valid DICOM formats. Third, only the dynamic series can be included when importing a study into CHOP-fMRU, and every image in the set must be present. If an image from another series is included or an image from the dynamic series is excluded, the user encounters an error. Finally, when the analysis has been performed, it is desirable to send the results to PACS. Unless the user has purchased an Interactive Data Language (IDL) license, the images may only be exported as JPEG files. These files must be packaged into DICOM files with the correct header data before they are sent to PACS using DICOM network protocol.

While these tasks could each be accomplished with any one of many available DICOM tools, some of the tasks are moderately complex, and performing them manually would be time consuming. Instead, an additional piece of software desigend solely to perform these functions as simply and efficiently as possible can be helpful. Such a solution was attempted at OHSU, and the resulting software, CHOP-fMRU Assistant, is presented here.

CHOP-fMRU Assistant is a program written in C# using the Grassroots DICOM (GDCM) library. Current features include automatic transfer syntax conversion, removal of many potentially incompatible data elements, easy exclusion of unwanted series, analysis of the dynamic series for missing images, and conversion of the results from CHOP-fMRU to DICOM files for upload to PACS. CHOP-fMRU Assistant is available at https://github.com/overtone1000/CHOP-fMRU_Assistant under an open-source license and is offered with an associated wiki to help other institutions who have encountered similar barriers while attempting to implement CHOP-fMRU.
Methods & Materials:
Results:
Conclusions: