14480
clinical study

Implementation of an AI-Driven PERT Workflow at a Large Academic Institution

Materials & Methods
This retrospective study evaluated the impact of an AI-based pulmonary embolism (PE) detection and notification system (Aidoc, Tel Aviv, Israel) on Pulmonary Embolism Response Team (PERT) activations. The AI algorithm continuously analyzed CT pulmonary angiograms (CTPAs) for peri-central thrombus burden and right ventricular strain (RV/LV ratio > 1.0). Positive detections automatically alerted the PERT team via a mobile platform integrating hemodynamic, imaging and laboratory data. Traditional pager-based activations remained active for comparison.

Results
From July 2024 to April 2025, 26,168 CTPAs were processed, identifying 1,532 positive PE studies. Of these, 383 met criteria for AI-driven PERT activation. Traditional activation yielded 552 cases; dual activation occurred in 247, while 136 were AI-only. RV/LV ratios were significantly higher in AI-activated cases versus traditional-only (1.39 ± 0.31 vs. 1.09 ± 0.21, p < 0.001). Procedural interventions occurred in 33.6% of dual-activation cases versus 9.7% of traditional-only (p < 0.001).

Conclusions
AI-integrated PERT activation enabled earlier identification of intermediate- to high-risk PE cases using radiologic and hemodynamic criteria. The hybrid AI-clinician workflow reduced activation delays and increased procedural interventions, demonstrating the potential of AI-assisted triage to optimize multidisciplinary PE management.

Disclaimer: This links to a third-party website, which may require credentials to review.

Explore the Latest AI Insights, Trends and Research