Automatic rib cage unfolding with CT cylindrical projection reformat in polytraumatized patients for rib fracture detection and characterization: Feasibility and clinical application

Abstract

ObjectivesTo assess the use of CT with unfolded cylindrical projection (UCP) for rib fracture detection and characterization. MethodsThe images from 60 polytraumatized patients who underwent whole body CT were evaluated for the presence and characterization of rib fractures (displaced or not, single or multiple). Two readers independently evaluated conventional CT images and UCP images in two readout sessions at least one month apart. All readouts were timed. A gold standard was established by two radiologists with 12 and 22 years of clinical experience based on the combined analysis of conventional and UCP reformats. ResultsUsing UCP, the mean evaluation time was 27%–54% shorter (P = 0.01 and <0.0001) while maintaining a comparable diagnostic performance (sensitivity and specificity of 68.4–79.1% and 99.5–99.6% for conventional reformats and 70.6–91.0% and 96.8–97.7% for UCP) and a good reproducibility (Kappa of 0.71). The multiple fracture detection ratio of UCP was similar to that of conventional reformats (>80%). There were more false positives and false negatives using UCP and displaced fractures were harder to characterize. ConclusionUCP yielded a diagnostic performance similar to that of conventional reformats for the detection of rib fractures with a good reproducibility and a noticeable reduction in evaluation time.