The Dose Index Registry (DIR) is a data registry that allows facilities to compare their CT dose indices to regional and national values. Information related to dose indices for all CT exams is collected, anonymized, transmitted to the ACR, and stored in a database. Institutions are then provided with periodic feedback reports comparing their results by body part and exam type to aggregate results. Data collected from the registry will be used to establish national benchmarks for CT dose indices.

Integrating the Healthcare Enterprise (IHE) is a global initiative that creates the framework for passing vital health information seamlessly. IHE, in collaboration with the ACR, has established a Radiation Exposure Monitoring (REM) Profile. This profile allows all vendors to collect and transmit the information related to CT dose indices and techniques in a similar format. The REM profile uses the DICOM CT Dose SR (PS 3.16 ftp://medical.nema.org/medical/dicom/2009/09_16pu.pdf). The CT Dose SR defines the data elements related to the specific dose information that is to be collected and describes the standard format in which the information should be stored. The DIR collects the dose information in the DICOM CT Dose SR and some additional information contained in the metadata in the DICOM header of the SR object. However, to maintain patient privacy, only limited information related to the patient (e.g., age in years, sex) is transmitted to the registry. Patient identifiers are removed at the facility before the data are transmitted to the DIR.

To participate in the registry, a facility enrolls in the DIR through the NRDR portal and is assigned a unique facility identifier. The ACR's TRIAD software is then installed on a PC at the participating facility. The CT scanner automatically sends the appropriate DICOM SR object to the PC for every CT exam performed on the scanner. TRIAD is used to anonymize and transmit data to the registry. Periodically, reports are provided to the facility through the NRDR portal comparing the facility's results to aggregate results by exam type. Comparisons to both regional and national data are provided. In addition, facilities can download their own data and generate reports and graphs from their data in real time.

Currently there is no standard for naming exam types across facilities. For example, an adult CT of the head without contrast might be labeled as "Head 1Brain_without (Adult)" at one facility and as "Head^1_HEAD_WO (Adult)" at another. To handle this problem, the DIR is implementing the RadLex Playbook. The Playbook is the list of standard-name procedures with a RadLex Playbook ID (RPID). The non-standard exam names from each facility are mapped to the Playbook so that accurate comparisons can be made.