59496 First Clinical Experience With a Real-Time Artificial Intelligence Assistance System for Neurointerventional Device Navigation in the United States

View at publisher Summarise with VIRAT
Authors: Joshua Jimenez, João Victor Sanders, Krishna C. Joshi, Marion Oliver, Kenichi Kono, Demetrius K. Lopes
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

INTRODUCTION: Neurointerventional procedures require simultaneous device navigation and rapid decisions under high cognitive load, increasing the risk of unintentional device migration and delayed recognition of critical events. Real-time artificial intelligence (AI) assistance may enhance procedural safety by providing automated device tracking and actionable notifications. METHODS: In a prospective, single-center study, 21 patients underwent neuroendovascular procedures using Neuro-Vascular Assist on biplane fluoroscopy. The system tracked guidewires, microcatheters, coils, filters, and stents, providing real-time auditory and visual notifications upon device entry or exit from predefined regions of interest (ROIs). Notifications were categorized as true positive (TP), false positive (FP), or false negative (FN). TPs prompting operator adjustments within 10 seconds were classified as clinically relevant (CR). RESULTS: The system operated without technical issues and did not extend procedure time. Across 21 procedures, 245 notifications were recorded (232 TPs, 13 FPs, 17 FNs), with 94.7% precision and 93.2% recall. Mean notifications per case were 11.7 ± 8.3. Among 59 TPs evaluated, 25 (42.4%) led to immediate operator action, indicating actionable value for procedural safety monitoring. CONCLUSIONS: This first U.S. clinical experience with real-time AI device tracking during neuroendovascular procedures demonstrates feasibility, high accuracy, and actionable notifications supporting procedural safety. AI tools may reduce cognitive burden and improve intra-procedural situational awareness in complex neurointerventions.

Fields of Study & Subjects

Artificial Intelligence in Healthcare and Education Intracranial Aneurysms: Treatment and Complications Cerebrovascular and Carotid Artery Diseases

Publication details

Journal / source
Neurosurgery
Volume
72
Issue
Supplement_1
Pages
305–305
Publication date
26 March 2026
Language
EN