New study shows Brainomix e-CTA® provides similar measures to experts and improves inter-rater reliability
OXFORD, England, June 21, 2019 /PRNewswire/ -- e-CTA®, developed by Brainomix, provides fast, consistent collateral assessments powered by AI and Deep Learning, based on CTA analysis. The recently published CATS study showed that e-CTA® provides a real-time and fully automated approach to collateral scoring, with the potential to improve consistency of image interpretation.
The study demonstrated the high sensitivity and specificity of e-CTA®, independent of clinician input. This is very important especially when selecting ischemic stroke patients presenting symptoms beyond 6 hours or with an uncertain time of onset, as reflected in the recent European Stroke Organisation guidelines for mechanical thrombectomy. Expert neuroradiology input is rarely available in real-time at most hospitals around the world, especially at primary stroke centres from which patients need to be transferred to a comprehensive centre. Even when specialist neuroradiology opinion is readily available, e-CTA® offers the opportunity to improve inter-rater reliability, therefore helping select the right patients that can benefit the most from mechanical thrombectomy.
Dr George Harston, Brainomix Chief Medical and Innovation Officer & Consultant Stroke Physician, Oxford University Hospitals NHSFT stated: "The importance of good collateral blood flow for patient outcome in acute stroke treatment is well recognised. However, even expert collateral scores commonly suffer from inter-rater error. In this exciting study Brainomix has shown once again the value of AI to augment human performance. In this study, e-CTA improved consistency between neuroradiologists when used as a decision support tool. Even when used as a standalone tool e-CTA produced similar scores compared to experts, opening up the opportunity for reliable collateral assessment for all patients regardless of which hospital they attend."
Read the full study here.