• Informatics
  • RadLex in Your Practice

  • RadLex enables numerous improvements in the clinical practice of radiology, from the ordering of imaging exams to the use of information in the resulting report. Developers of educational tools can use RadLex to organize their materials and make their contents more definitive. Reporting templates developed by the RSNA Reporting Committee use RadLex terms in their content, resulting in clearer and more consistent reports. Clinical trials investigators can use RadLex terms to "tag", index, search and analyze radiology-related medical data.
    Some specific uses of RadLex terminology include:

    • Automatic order entry decision support. Because the names of imaging exams are described in consistent language, choosing the appropriate template is easy and intuitive. The applicability of appropriateness criteria developed by the ACR and others can be determined automatically.

    • Vendor independent "protocoling" of complex imaging exams. Imaging exam protocols for CT and MR exams can be specified using vendor independent language. Consistent names for imaging exams and procedure steps (e.g., radiographic view, CT sequence, MR series) are used throughout the radiology practice.

    • Reliable PACS display layouts. Because MR and CT sequences have consistent names, PACS display layouts can be defined in a consistent language used by all PACS vendors. When an imaging exam is launched on a PACS workstation, the images are arranged automatically according to the preferences of the user. As a result, display layouts are more reliable, and can be transferred from one PACS system to another.

    • No need to re-dictate lengthy imaging technique sequences. Because imaging devices use consistent language to describe imaging procedure steps, a description of those steps can be incorporated automatically into the radiology report and need not be re-dictated.

    • Improved speech recognition accuracy. Because the exam descriptions are explicitly linked to the body site imaged and the modality used, speech recognition systems use this linkage information to improve recognition accuracy.

    • Speech-enabled structured reporting to satisfy regulatory requirements. For key clinical measures, "blanks" in report templates can be filled in from speech-triggered menus containing standardized terms. As a result, pay-for-performance and other quality measures can be reported routinely without manual searches.

    • Real-time decision support for the radiologist. Because standardized terms are associated with radiology reports, these terms can trigger decision support tools for the radiologist. Decision support systems automatically retrieve case-relevant information in real time, such as checklists for image features to seek, additional differential diagnoses, or information from PubMed, the Internet, and proprietary decision support databases.

    • Rapid teaching file creation. Standardized terms associated with the radiology report, whether assigned by the radiologist or created automatically from text-processing algorithms, facilitate the automatic indexing and organization of radiology teaching files.

    • Accurate report search and data mining. Because standardized terms are associated with each report, report databases can be mined to address regulatory requirements, meet pay for performance objectives, or measure performance improvement goals. For example, standard lexicons can be used to solve data mining problems due to synonyms (a search for "renal stone" does not retrieve reports mentioning "kidney stone" or "urolithiasis"), negation (a search for "ectopic pregnancy" overwhelmingly yields reports stating "there is no evidence of ectopic pregnancy"), and inheritance (a search for "lung" and "cancer" does not retrieve reports describing "adenocarcinoma in the right upper lobe"). Report data can readily identify important trends and satisfy specific objectives.

    Users and developers of imaging-related HIT systems and other applications that can benefit from radiology-specific coded terminology are invited to adopt RadLex and take advantage of the major advances in the practice of radiology that it makes possible.