North Texas GME Research Forum 2024
Files
Download Poster or Presentation (266 KB)
Division
North Texas
Hospital
Medical City Arlington
Specialty
Emergency Medicine
Document Type
Poster
Publication Date
2024
Keywords
quality improvement, stroke, AI, artificial intelligence
Disciplines
Emergency Medicine | Medicine and Health Sciences | Quality Improvement
Abstract
Introduction: Endovascular therapy (EVT) in the form of mechanical thrombectomy is the mainstay of treatment for acute large vessel occlusion (LVO) ischemic stroke, but its efficacy is highly time sensitive. It is crucial that stroke centers continue to implement process improvements that aim to streamline stroke workflow. Viz.AI is a platform that uses artificial intelligence to automatically detect LVOs with computed tomography (CT) imaging. It provides immediate access to the CT images as well as a platform for centralized communication through the mobile application. We sought to determine if the implementation of Viz.AI at our comprehensive stroke center improved stroke workflow and metrics. Methods: We conducted a retrospective review of all LVO stroke cases that underwent EVT at our facility from June 2020 through December 2022. Transfers and inpatient strokes were excluded. We compared periods before and after implementation of Viz.AI. The primary outcome was mean time of arrival in the emergency department to arrival at interventional radiology suite (EDIR). Rates of substantial reperfusion following EVT (modified thrombolysis in cerebral infarction score of 2C/3) were compared as a secondary outcome. Data was analyzed using t-test. Results: There were a total of 78 patients with LVO stroke who met inclusion criteria and underwent EVT from June 2020 to December 2022. There were 35 cases from June 2020 through June 2021 (pre-Viz.AI group) and 43 cases from July 2021 through December 2022 (post-Viz.AI group). Implementation of Viz.AI resulted in 16 min reduction in EDIR times (125 vs 109 minutes, P=0.09). There was no significant difference in rates of substantial reperfusion with EVT between groups (74% vs 71%). Conclusions: In patients with acute LVOs, there was no statistically significant improvement following implementation of Viz.AI at our facility.
Original Publisher
HCA Healthcare Graduate Medical Education
Recommended Citation
Bryant, Cody; Womack, Brandon; Patel, Mason; and Sharma, Sanjay, "Implementation of Viz.ai Augmented Intelligence Software to Reduce Door to Puncture Times and Patient Outcomes at a Comprehensive Stroke Center" (2024). North Texas GME Research Forum 2024. 14.
https://scholarlycommons.hcahealthcare.com/northtexas2024/14