Detection of ischemic changes on baseline multimodal computed tomography: expert reading vs. Brainomix and RAPID software

Purpose: The aim of the study was to compare the assessment of ischemic changes by expert reading and available automated software for non-contrast CT (NCCT) and CT perfusion on baseline multimodal imaging and to demonstrate accuracy for final infarct prediction.

Methods: Early ischemic changes were measured by ASPECTS on baseline neuroimaging of consecutive patients with anterior circulation ischemic stroke, assessed on NCCT by two experienced raters, on NCCT by e-ASPECTS, and visually on derived CT perfusion maps (CBF<30%, Tmax>10s). Accuracy was calculated against final ischemic changes on 24-hour follow-up for each ASPECTS region (sensitivity, specificity, PPV, NPV), with a subanalysis for successful recanalization.

Results: Of 263 patients, 81 met inclusion criteria; median baseline ASPECTS was 9 for all modalities. Accuracy was 0.76 for e-ASPECTS, 0.79 for consensus, 0.82 for CBF<30%, and 0.80 for Tmax>10s. Results did not differ in patients with and without successful recanalization.

Conclusion: This study demonstrated high accuracy for the assessment of ischemic changes by different CT modalities, with the best accuracy for CBF<30% and Tmax>10s. Automated software has the potential to improve the detection of ischemic changes.