Abstract
Purpose
The purpose was to compare performance of diagnostic workstation monitors and the Apple iPad 2 (Cupertino, CA) in interpretation of emergency computed tomography (CT) brain studies.
Methods
Two experienced radiologists interpreted 100 random emergency CT brain studies on both on-site diagnostic workstation monitors and the iPad 2 via remote access. The radiologists were blinded to patient clinical details and to each other's interpretation and the study list was randomized between interpretations on different modalities. Interobserver agreement between radiologists and intraobserver agreement between modalities was determined and Cohen kappa coefficients calculated for each. Performance with regards to urgent and nonurgent abnormalities was assessed separately.
Results
There was substantial intraobserver agreement of both radiologists between the modalities with overall calculated kappa values of 0.959 and 0.940 in detecting acute abnormalities and perfect agreement with regards to hemorrhage. Intraobserver agreement kappa values were 0.939 and 0.860 for nonurgent abnormalities. Interobserver agreement between the 2 radiologists for both diagnostic monitors and the iPad 2 was also substantial ranging from 0.821-0.860.
Conclusions
The iPad 2 is a reliable modality in the interpretation of CT brain studies in them emergency setting and for the detection of acute and chronic abnormalities, with comparable performance to standard diagnostic workstation monitors.
Get full access to this article
View all access options for this article.