A Multi-institutional Data Harvesting (MIDH) method for retrospective, longitudinal observation of medical imaging utilization and reporting for computed tomography pulmonary angiograms (CTPA) (n=40,037) was conducted across 13 academic sites. Compared two 70-day observational periods, namely (i) a pre-pandemic control period from Nov. 25, 2019 to Feb. 2, 2020, and (ii) a period during the early COVID-19 pandemic from March 8, 2020 through May 16, 2020. Natural language processing (NLP) on final radiology reports served as the ground truth for identifying positive PE cases.
Fewer CTPA exams were performed during the early COVID-19 pandemic than during the pre-pandemic period (9,806 vs. 12,106). The PE positivity rate was significantly higher during the early COVID-19 pandemic than during the pre-pandemic period (11.6 vs. 9.9%, p < 10−4). NLP accuracy of 98% based on a manual review of 2,400 radiology reports
The positivity of PE was significantly higher during the COVID era and fewer CTPA exams were performed. This MIDH method showed the successful scalability and ability of the Aidoc solution to overcome typical challenges of multi-institution data silos while maintaining a high level of accuracy in PE classification.
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.
