{
  "abstract": "Background Rapid identification of intracranial vessel occlusions (VOs) is critical for stroke triage and endovascular care. Workflow constraints and interpretive variability may contribute to undiagnosed VOs on CT angiography (CTA). This study evaluated the impact of an AI-augmented CTA workflow on VO detection in routine clinical practice.Methods We included 1,472 consecutive adult CTA studies performed over five months (November 2023 through April 2024) in a large, integrated, multisite healthcare system. In this population, the AI algorithm was not yet implemented during the initial read. Unaided radiologist diagnoses were extracted from the associated radiology report using natural language processing. A commercially available, FDA-cleared AI algorithm (Aidoc, Tel Aviv, Israel) then retrospectively evaluated all CTA studies and identified those with suspected VOs. Cases identified as negative for VO in the radiology report but positive by the AI algorithm underwent adjudication for ground truth by two interventional neuroradiologists. Enhanced detection was defined as AI-identified VOs not reported during initial interpretation.Results Initial radiologist interpretation identified 83 VO positive cases (5.6%). The AI algorithm flagged 20 additional cases, of which 15 were confirmed true positives, representing an 18.1% (15/83) increase in detected VOs. Newly identified cases included 8 medium vessel occlusions (1 V4, 3 M2, 3 M3, 1 P2) and 7 large vessel occlusions (2 M1 and 5 P1). All AI false positives were localized to the V4 vertebral artery. Annualized extrapolation suggests AI-assisted triage could identify approximately 38 additional VOs per year.Conclusion Integration of AI into CTA workflows improves detection of intracranial VOs and enhances case prioritization without disrupting radiologist autonomy. AI-driven triage represents a scalable workflow innovation to support timely endovascular evaluation and potentially decrease stroke mortality.Disclosures J. Milburn: 2; C; AI Doc. P. Gulotta: None. V. Loving: None. A. Toshav: None.",
  "authors": [
    {
      "affiliations": [
        "Radiology, Ochsner Medical Center, Jefferson, LA"
      ],
      "name": "J Milburn"
    },
    {
      "affiliations": [
        "Radiology, Ochsner Medical Center, Jefferson, LA"
      ],
      "name": "P Gulotta"
    },
    {
      "affiliations": [
        "Ochsner Medical Center, Jefferson, LA"
      ],
      "name": "V Loving"
    },
    {
      "affiliations": [
        "Radiology, Ochsner Medical Center, Jefferson, LA"
      ],
      "name": "MC Morvant"
    },
    {
      "affiliations": [
        "Radiology, Ochsner Medical Center, Jefferson, LA"
      ],
      "name": "A Toshav"
    }
  ],
  "title": "E-359 AI-enhanced CTA triage improves detection of intracranial vessel occlusions in routine clinical workflow",
  "uid": "cb25976d-34ee-5a1f-8e46-5e8637e1a5ee"
}