Clinical neuroradiology
September 21, 2026 · via Europe PMC
Clinical Implementation and Performance of Real-Time AI System for Device Tracking in Neurointervention: First U.S. Evaluation of Neuro-Vascular Assist.
Sanders JV, Jimenez J, Oliver M, Kono K, Joshi KC, Lopes D.
AI summary — for doctors, generated from the abstract only
A real-time AI device-tracking system for neurointervention ran without workflow disruption in its first U.S. use, with about 95% precision and 93% recall for device notifications.
What was studied
Single-center feasibility study of Neuro-Vascular Assist integrated with a biplane angiography system across 21 consecutive neurointerventional cases, scoring notification accuracy and operator response.
Key findings
- 232 true positives, 13 false positives, 17 false negatives: precision 94.7%, recall 93.2%
- No procedural delays or system failures affecting notifications
- 42.4% of 59 evaluated true-positive alerts prompted an observable operator response within 10 s (10.2% of all notifications)
Clinical relevance
This supports technical feasibility of real-time AI device tracking in the angio suite, but the small, single-center sample says nothing yet about effects on safety or outcomes.
Who should care
Limitations
Very small single-center series (21 cases) with no clinical outcome or comparator data; operator response assessed in only a subset of alerts.
AI device-tracking alerts appear accurate and workable in early use, but their clinical value is still unproven.
This is an AI-generated summary of the published abstract, not a clinical recommendation. It may not reflect the full text or subsequent updates — verify against the original source before acting on it.
Abstract
Purpose
To evaluate the first U.S.-based implementation, notification accuracy, and temporal operator response of Neuro-Vascular Assist.
Methods
In a single-center study, 21 consecutive neurointerventional cases were performed using the AI system integrated with a biplane angiographic system. Procedures were classified as "Standard", "Coil", "Filter", or "Venous". Neuro-Vascular Assist tracked devices including guidewires, stents, coils, and filters in real-time. Visual and audio notifications were made based on device entry and exit in a determined region of interest. We classified these notifications as true positives (TP), false positives (FP), and false negatives (FN), and further assessed for temporal operator response (TOR).
Results
The AI system was successfully integrated into the workflow without procedural delays or system failures affecting notification delivery. A total of 232 TPs, 13 FPs, and 17 FNs were recorded, resulting in an overall precision of 94.7% and recall of 93.2%. Among the 59 TP notifications evaluated for TOR, 25 (42.4%) were followed by an observable operator response within 10 s. These 25 positive TOR events represented 10.2% of all 245 notifications.
Conclusion
Neuro-Vascular Assist was successfully implemented and demonstrated preliminary accuracy and feasibility in a U.S.-based neurointerventional setting. Larger prospective studies are warranted to determine whether real-time AI-assisted device tracking influences procedural decision-making, workflow, safety, or clinical outcomes.
DOI
10.1007/s00062-026-01723-8
PMID
42766128
Research metadata (titles, abstracts, authors, publication dates) is sourced from Europe PMC, which aggregates content from PubMed/MEDLINE, PubMed Central, and other providers under the U.S. National Library of Medicine (NLM/NCBI). NLM does not claim copyright on abstracts; the original journal publishers or authors may hold rights to the full articles. AI-generated summaries are derived only from the abstract, never the full text, and are always labeled as AI-generated.
Want more like this?
Browse more from the feed. A free account unlocks the full feed, saving and communities.
Back to the research feed