Amit D. Kalaria MD, Presenter: Nothing to Disclose

Ross Warren Filice MD, Abstract Co-Author: Nothing to Disclose

Regular comparison of preliminary to final reports is a critical part of radiology resident and fellow education as prior manual comparisons have documented substantial preliminary to final discrepancies. Unfortunately, there are many barriers to this comparison: high study volume; overnight rotations without an attending; the ability to finalize reports remotely; the subtle nature of many changes; and the loss of the preliminary report after finalization.

We receive a real-time Health Level 7 (HL7) feed from our Radiology Information System (RIS) (Siemens). Our Mirth Connect HL7 engine (Mirth Corporation) filters radiology report messages and parses report data into a MySQL (Oracle) database separated by preliminary and final status. A Bash (GNU) script queries all preliminary reports over a desired time period, finds the associated final report, evaluates for differences, compiles positive difference reports by radiologist, and emails these reports automatically.

Our system automatically compiles and emails a weekly summary of report differences for all residents and fellows. Differences between preliminary and final report are clearly highlighted with links to the associated study in PACS for efficient review and learning. This provides an easy way to review changes to preliminary reports with immediate access to the associated images. We hypothesize that this will improve our education and learning experience and may also decrease discrepancy rates over time. A comparison of discrepancy rates prior to implementation of this system with prospective discrepancy rates will be presented to quantify impact. Subjective resident and fellow survey results regarding opinions of the weekly report and how it affects their education and learning experience will also be summarized across PGY levels. We will continue to refine our discrepancy algorithm to try and more intelligently distinguish significant differences and prioritize the presentation order in the weekly reports.

We believe this system will improve our department education and learning experience and will hopefully reduce future radiology report discrepancy rates.

Kalaria, A, Filice, R, Comparison-Bot: An Automated Preliminary-Final Report Comparison System.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13012484.html