This single-center, IRB-approved observational study evaluated the impact of an FDA-cleared AI algorithm (Aidoc) on detecting suspected incidental pulmonary embolisms (iPEs) in chest CT scans performed for non-PE indications.
A total of 5,298 scans were reviewed and divided into two cohorts: cohort 1 consisted of 1,964 patients whose radiology reports were generated before AI implementation, while cohort 2 included 3,334 patients whose scans were analyzed with AI assistance. Discrepancies between radiology reports and suspected AI findings were reviewed by two thoracic imaging specialists, with the expert review serving as the reference standard.
Metrics such as sensitivity, specificity, prevalence, PPV and NPV were calculated for each cohort. Survival analysis was conducted using Kaplan-Meier curves and Cox proportional hazards models to assess short-term outcomes.
In cohort 1 (pre-AI), the prevalence of confirmed iPEs was 2.2% (42/1,964), with 50% of cases overlooked by radiologists. The AI system achieved a sensitivity of 95%, specificity of 99%, PPV of 69% and NPV of 99%. In cohort 2 (post-AI), the prevalence of confirmed iPEs was 1.7% (56/3,334), with 7.1% of cases overlooked by radiologists, demonstrating a significant reduction in overlooked cases (P < 0.001). AI performance remained high with a sensitivity of 90%, specificity of 99%, PPV of 95% and NPV of 99%. Notably, 61% of missed iPEs occurred in the right lower lobe. No statistically significant impact on survival was observed.
The implementation of an AI algorithm reduced overlooked cases of iPEs 7.1%, significantly enhancing radiologists’ diagnostic accuracy. Radiologists are advised to focus on the right lower lobe, where most overlooked cases of iPEs occur.
* The effect of using the Briefcase-Triage to reduce missed iPE on Chest CT was done as part of clinical investigation for research purposes.
The Briefcase-Triage is indicated to assist hospital networks and appropriately trained medical specialists in workflow triage by flagging and communication of suspect cases of iPE pathologies. For the full list of indications for use and feature availability, please refer to the User Guide.
Aidoc experts, customers and industry leaders share the latest in AI benefits and adoption.
Explore how clinical AI can transform your health system with insights rooted in real-world experiences.
Learn how to go beyond the algorithm to develop a scalable AI strategy and implementation plan.
Explore how Aidoc can help increase hospital efficiency, improve outcomes and demonstrate ROI.
