The use of artificial intelligence to improve detection of acute incidental pulmonary emboli.

Background: Incidental pulmonary emboli (IPE) are frequently overlooked by radiologists. Artificial intelligence (AI) algorithms have been developed to aid detection of pulmonary emboli.

Objectives: To measure diagnostic performance of AI compared with prospective interpretation by radiologists.

Methods: A commercially available AI algorithm was used to retrospectively review 14 453 contrast-enhanced outpatient computed tomography examinations of the chest, abdomen, and pelvis in 9171 patients, where pulmonary embolism was not clinically suspected. Natural language processing searches of reports identified IPE detected prospectively. Thoracic radiologists reviewed all cases read as positive by AI or natural language processing to confirm IPE and assess the most proximal level of clot and overall clot burden. A total of 1400 cases read as negative by both the initial radiologist and AI were rereviewed to assess for additional IPE.

Results: Radiologists prospectively detected 218 IPE, and AI detected an additional 36 unreported cases. AI missed 30 cases of IPE detected by the radiologists and had 94 false positives. For 36 IPE missed by the radiologists, median clot burden was 1, and 19 were solitary segmental or subsegmental. For 30 IPE missed by AI, 1 case had large central emboli, and the others were small with 23 solitary subsegmental emboli. Radiologist rereview of 1400 examinations interpreted as negative found 8 additional cases of IPE.

Conclusion: Compared with radiologists, AI had similar sensitivity but reduced positive predictive value. Our experience indicates that the AI tool is not ready to be used autonomously without human oversight, but a human observer plus AI is better than either alone for detection of IPE.