Impact of Artificial Intelligence Assistance on Chest CT Interpretation Times: A Prospective Randomized Study.
Researchers
Journal
Modalities
Models
Abstract
Background: Deep learning-based convolutional neural networks have enabled major advances in development of artificial intelligence (AI) software applications. Modern AI applications offer comprehensive multiorgan evaluation. Objective: To evaluate the impact of an automated AI platform, integrated into clinical workflow for chest CT interpretation, on radiologists’ interpretation times when evaluated in a real-world clinical setting. Methods: In this prospective single-center study, a commercial AI software solution was integrated into clinical workflow for chest CT interpretation. The software provided automated analysis of cardiac, pulmonary, and musculoskeletal findings, including labeling, segmenting, and measuring normal structures as well as detecting, labeling, and measuring abnormalities. AI-annotated images and autogenerated summary results were stored in the PACS and available to interpreting radiologists. A total of 390 patients (204 female, 186 male; mean age, 62.8±13.3 years) who underwent outpatient chest CT between January 19, 2021 and January 28, 2021 were included. Scans were randomized using 1:1 allocation between AI-assisted and non-AI arms and were clinically interpreted by one of three cardiothoracic radiologists (65 scans per arm per radiologist; total of 190 scans per arm), who recorded interpretation times using a stopwatch. Findings were categorized based on review of report impressions. Interpretation times were compared between arms. Results: Mean interpretation times were significantly shorter in the AI-assisted than in the non-AI arm for all three readers (289±89 vs 344±129 seconds, p<.001; 449±110 vs 649±82 seconds, p<.001; 281±114 vs 348±93 seconds, p=.01) and for readers combined (328±122 vs 421±175 seconds, p<.001). For readers combined, the mean difference was 93 seconds (95% CI, 63-123 seconds), corresponding with a 22.1% reduction in the AI-assisted arm. Mean interpretation time was also shorter in the AI-assisted arm compared with the non-AI arm for contrast-enhanced scans (83 seconds), non-contrast scans (104 seconds), negative scans (84 seconds), positive scans without significant new findings (117 seconds), and positive scans with significant new findings (92 seconds). Conclusion: Cardiothoracic radiologists exhibited a 22.1% reduction in chest CT interpretations times when having access to results from an automated AI support platform during real-world clinical practice. Clinical Impact: Integration of the AI support platform into clinical workflow improved radiologist efficiency.