CINA-iPE Deep Learning Algorithm Accurately Identifies Incidental PE on CT Scans

Background: Prompt detection of pulmonary embolism (PE) is critical to reduce morbidity and mortality. Dedicated CT pulmonary angiography (CTPA) demonstrates high diagnostic accuracy, and deep learning algorithms have shown promise in detecting PE on CTPA. Whether these methods can identify PE on routine contrast-enhanced CT (CECT) studies remains less well studied. Objective: To evaluate the performance of a commercially available deep learning algorithm in detecting PE on routine CECT scans not performed with a dedicated CTPA protocol. Design/Participants: Retrospective, multinational multicenter study analyzing adult patients (aged ≥18 years) undergoing routine CECT scans between January 2020 and December 2023 for indications other than suspected PE. Methods: CECT scans (slice thickness ≤2.5 mm, axial acquisitions only) were acquired on 39 different scanner models. Artificial intelligence (AI) analysis was performed using CINA-iPE v1.0 software (Avicenna.AI, France). Two experienc more...