Mobile device for thrombolysis decisions for telestroke

Abstract

Aim: This study compares the reliability of brain CT interpretations performed using a diagnostic workstation and a mobile tablet computer in a telestroke context. Methods: A factorial design with 1,452 interpretations was used. Reliability was evaluated using the Fleiss’ kappa coefficient on the agreements of the interpretation results on the lesion classification, presence of imaging contraindications to the intravenous recombinant tissue-type plasminogen activator (t-PA) administration, and on the Alberta Stroke Program Early CT Score (ASPECTS). Results: The intra-observer agreements were as follows: good agreement on the overall lesion classification (κ= 0.63, p <0.001), very good agreement on hemorrhagic lesions (κ= 0.89, p <0.001), and moderate agreements on both without acute lesion classification and acute ischemic lesion classification (κ= 0.59 and κ= 0.58 respectively, p <0.001). There was good intra-observer agreement on the dichotomized-ASPECTS (κ= 0.65, p <0.001). Conclusions: The results of our study allow us to conclude that the reliability of the mobile solution for interpreting brain CT images of patients with acute stroke was assured, which would allow efficient and low-cost telestroke services.