Abstract
This study analyzes if artificial intelligence (AI) technology is as reliable as neuroradiologist interpretation to identify patients who are candidates for endovascular thrombectomy (EVT) in the setting of acute ischemic stroke. The AI technology utilized in this study is called RAPID. This study also examines limitations of RAPID and aims to ascertain the degree to which RAPID can be used to triage patients for EVT in the acute setting. This is an Institutional Review Board approved, retrospective study of 857 patients presenting to our institution from 1/1/2018 to 11/1/2019 with symptoms of acute stroke. Inclusion criteria included patients who entered our institutional stroke protocol and underwent CT, CTA and CTP scans interpreted by both RAPID and radiologists. Hospital records were also obtained to determine which patients were transferred for EVT and their post-transfer outcomes. A total of 294 patients satisfied the inclusion criteria. Overall, RAPID CTP analysis showed significant sensitivity (90.7%), specificity (92.5%) and negative predictive value (100%) for identifying EVT-eligible patients when using radiologist interpretation as the gold standard. RAPID CTA showed significant specificity (95.9%) and negative predictive value (94.5%) as well. There was also a significant association between RAPID interpretation and neuroradiologist interpretation of CTA and CTP studies. The high specificity and 100% negative predictive value (NPV) of RAPID suggests that this technology is a reliable screening tool to accurately rule out patients who are not EVT candidates. All patients who had a negative RAPID scan did not require transfer for EVT, and upon review of these cases by a current neuroradiologist, the decision to not transfer those patients was deemed accurate. Correctly identifying patients who are not candidates for EVT is a vital step in the management of patients with stroke and enables providers to more quickly begin the next phases of treatment. If the use of AI technology as a screening tool can be validated, such technology can be incorporated into hospital protocols to potentially expedite patient care. Furthermore, AI technology can be most impactful in the setting of community hospitals, which may lack 24-hour neuroradiologist availability. Overall, this study suggests that RAPID is a reliable screening tool to rule out stroke patients who are not candidates for EVT.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
