Les Folio DO, Presenter: Nothing to Disclose

Catherine McHugh, Abstract Co-Author: Nothing to Disclose

The intent of this project is to create a graphical tool of thoracic imaging patterns to serve as a high-level review for teaching and reference purposes. It is hypothesized that the tabular display, using an Excel spreadsheet, acts as an “aerial map” to improve comprehension and recognition of disease correlates of various patterns. Additional projects will apply the algorithm as a web-based teaching tool utilizing image comparisons along the chosen algorithm pathways. This algorithm may lay the groundwork for neural networks and web image searching, similar to fingerprint recognition and matching.

The tabular spreadsheet was created using Treeplan (Decision Support Services; SF, CA) commonly used for decision analysis and algorithm generation. The overall algorithm was integrated into the chest curriculum at the Uniformed Services University of the Health Sciences (USUHS) and well accepted in resident presentations. The Audience Response System (ARS) is used to measure success for interactive quizzes. In addition, questions were asked about the perceived value of the algorithm for comprehension and retention of chest imaging. The recorded responses were reviewed retroactively over the course of the academic year of fourth-year students. The ARS used is a commercial product called TurningPoint® (Turning Technologies, LLC).

Retrospective reviews of the ARS satisfaction questions of perceived teaching value are consistently favorable for using a standard approach. Retrospective review of interactive feedback in ARS revealed that 92 % (n = 74) students felt the algorithm helped them understand chest imaging. Students and residents that claim to be graphically centered learners are especially appreciative of such high-level tools.

The Chest Pattern Algorithm has promise in teaching students of radiology at all levels and may act as a framework for future web-based and automated diagnostic assistance using neural networks and pattern recognition. The algorithm may help students of radiology better comprehend chest imaging by mapping out pattern/ disease correlates.

The teaching algorithm serves as a diagnostic decision tool in the clinical setting.

Folio, L, McHugh, C, Thoracic Imaging Pattern: Disease Map.  Radiological Society of North America 2006 Scientific Assembly and Annual Meeting, November 26 - December 1, 2006 ,Chicago IL. http://archive.rsna.org/2006/4434145.html