• Maria Gabriela Figueiro Longo, Pieter Vuylsteke, Can Ozan Tan, Joris A M Soons, Laura Avery, Marc D Succi, Vinit Baliyan, Ali Pourvaziri, Limin Xu, Daniel Chonde, Michael Lev, Jeroen Cant, and Rajiv Gupta.
• Massachusetts General Hospital, Radiology Department, Boston, USA. Electronic address: mfigueirolongo@mgh.harvard.edu.
• Am J Emerg Med. 2022 Dec 1; 62: 626862-68.

PurposeTo investigate the accuracy and total assessment time (TAT) of the "All-in-one" (AIO)-window/level setting for whole-body computed tomography (CT) image compared to multiple tissue-specific window/level settings conventionally used for detection of traumatic injuries.MethodContrast-enhanced chest, abdomen, and pelvic CT scans of 50 patients who presented to our emergency department (ED) for major trauma were retrospectively selected. In a simulation of a "wet read" performed at the CT scanner console, 6 readers with different levels of experience had up to 3 min to describe any traumatic finding identified on the CTs. The readers reviewed each patient in two different sessions separated by a washout period to suppress any recall bias from one session to the next. Each scan was reviewed once using the AIO-window/level setting and another time using the conventional bone, lung, and soft tissue window/level display settings, in a randomized order. The CT reports were used as reference standard.ResultsOverall, there was no statistically significant difference in the assessment accuracy of the review based on the AIO or the conventional window/level settings (0.89 ± 0.09 vs 0.90 ± 0.08). Using the AIO-window/level settings, TAT was 14.3 s faster when compared with the conventional window/level settings (2.33 ± 0.63 vs 2.57 ± 0.51 min; p < 0.001).ConclusionsIn a time-delimited image review, similar diagnostic accuracy was reached faster using the AIO vs the conventional window/level settings. When providing a "wet read" at the CT console, the ability to identify traumatic injury using a single AIO-window/level may help expedite patient management.Copyright © 2022. Published by Elsevier Inc.

8 keywords...

hide…