Evaluated the performance of an artificial intelligence algorithm (AIA, Aidoc) for detecting incidental pulmonary embolism (iPE) in cardiac CT angiography (CCTA) scans. A total of 1,534 CCTA scans from 2021 to 2023 were analyzed. The AIA’s findings were compared to radiology reports reviewed through a natural language processing (NLP) system.
Despite a focused field of view in CCTA, iPE is detected in about 1% of scans. The AIA identified 27 scans as positive for PE. Of these, 22 were confirmed by a second review, and five were negated. Notably, the AIA detected 45.5% (10/22) of the confirmed cases that were missed by initial radiologists, all segmental or subsegmental PEs (p<0.05). The AIA demonstrated high sensitivity (100%), specificity (99.6%), accuracy (99.6%), positive predictive value (81.4%) and negative predictive value (100%).
The AIA is effective in identifying iPE in CCTA scans and could improve diagnostic accuracy and lead to earlier detection and treatment.
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.
