How reading imaging exams on a mobile device affects accuracy

Time is brain, as the famous saying goes, when treating acute stroke patients. And radiologists can often save time for those patients by reading their imaging studies on mobile devices—smartphones or laptop computers, for instance—instead of a traditional workstation.

But does interpreting examinations on such devices impact diagnostic accuracy? Researchers addressed that very question, sharing their findings in the Journal of the American College of Radiology.

“The objective of this study was to assess the accuracy of the interpretations made with these devices,” wrote Antonio J. Salazar, PhD, University of Los Andes in Bogotá, Columbia, and colleagues. “We hypothesized that they have equivalent accuracy to medical workstations for head CT interpretations in acute stroke patients.”

The study included data from 188 acute stroke patients who underwent head CT examinations at a single hospital from 2013 to 2018. More than 47% of those patients were male, and the mean patient age was 71.3 years old. Four neuroradiologists participated, acting as observers, and a workstation including DICOM-compliant viewer software was treated as the study’s reference. A smartphone was the first mobile alternative, and a laptop computer was the second mobile alternative.

All four radiologists read all examinations using the reference method, first mobile alternative and second mobile alternative. Interpretations occurred over the course of one year, and the order of examinations was different for each participating neuroradiologist.

The ROC curves were “very similar” for all three reading systems, the authors observed, suggesting “similar observer performance with different specificities and sensitivities.”

And looking at reading times, the workstation was associated with a mean reading time of 114 seconds, smartphones had a mean reading time of 143 seconds and the laptops had a reading time of 130 seconds.

“This difference was negligible compared with a neuroradiologist’s commute time to the hospital in our city,” the authors wrote.