14477
clinical study

Implementation of an AI-Enabled PERT Workflow for Dynamic Risk Stratification in Pulmonary Embolism Across a Multi-Hospital Health System

Materials & Methods
This six-month, multi-hospital study (Sept. 2024–March 2025) evaluated the deployment of an AI-driven Pulmonary Embolism Response Team (PERT) workflow (Aidoc, Tel Aviv, Israel) integrating CT-based PE detection with electronic health record (EHR) data for real-time risk stratification. The platform automatically extracted RV/LV ratios, vitals and biomarkers (lactate, troponin, BNP, vasopressor use) to assign patients into institutional-specific risk tiers: high, intermediate or low. Automated alerts were triggered when patient parameters crossed severity thresholds within 24 hours of imaging.

Results
Among 73,908 CT studies analyzed, 1,394 (1.9%) were positive for PE, including 36% incidental and 64% suspected cases. Of these, 59% were categorized as intermediate-low, intermediate-high, or high risk, with dynamic reclassification observed in patients whose vitals or labs worsened. The AI-driven mobile platform enabled real-time alerts and targeted PERT activation, facilitating earlier multidisciplinary engagement and data-driven treatment prioritization.

Conclusions
The AI-enabled PERT workflow successfully automated PE detection and continuous risk stratification across a large health system. Integration of imaging AI with EHR data enabled dynamic, real-time monitoring and improved clinical responsiveness, supporting broader implementation of intelligent, alert-based PE management pathways.

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

Explore the Latest AI Insights, Trends and Research