Abstract
BackgroundArtificial intelligence (AI) in diagnostic radiology is undergoing rapid development. Its potential utility to improve diagnostic performance for cardiopulmonary events is widely recognized, but the accuracy and precision have yet to be demonstrated in the context of current screening modalities. Here, we present findings on the performance of an AI convolutional neural network (CNN) prototype (AI-RAD Companion, Siemens Healthineers) that automatically detects pulmonary nodules and quantifies coronary artery calcium volume (CACV) on low-dose chest CT (LDCT), and compare results to expert radiologists. We also correlate AI findings with adverse cardiopulmonary outcomes in a retrospective cohort of 117 patients who underwent LDCT.MethodsA total of 117 patients were enrolled in this study. Two CNNs were used to identify lung nodules and CACV on LDCT scans. All subjects were used for lung nodule analysis, and 96 subjects met the criteria for coronary artery calcium volume analysis. Interobserver concordance was measured using ICC and Cohen’s kappa. Multivariate logistic regression and partial least squares regression were used for outcomes analysis.ResultsAgreement of the AI findings with experts was excellent (CACV ICC = 0.904, lung nodules Cohen’s kappa = 0.846) with high sensitivity and specificity (CACV: sensitivity = .929, specificity = .960; lung nodules: sensitivity = 1, specificity = 0.708). The AI findings improved the prediction of major cardiopulmonary outcomes at 1-year follow-up including major adverse cardiac events and lung cancer (AUCMACE = 0.911, AUCLung Cancer = 0.942).ConclusionWe conclude the AI prototype rapidly and accurately identifies significant risk factors for cardiopulmonary disease on standard screening low-dose chest CT. This information can be used to improve diagnostic ability, facilitate intervention, improve morbidity and mortality, and decrease healthcare costs. There is also potential application in countries with limited numbers of cardiothoracic radiologists.
Highlights
Artificial intelligence (AI) in diagnostic radiology is undergoing rapid development
The purpose of this study was to investigate the performance of a fully automated AI convolutional neural network (CNN, a multi-layered machine learning algorithm which utilizes multiple hidden layers and sequential output patterns that excel at image) in simultaneously detecting solid pulmonary nodules and quantifying calcium volume (CACV) on routine low-dose chest CT (LDCT) scans of the chest when compared against expert radiologists
Clinical attributes, risk factors, and univariate statistics Demographics of patients evaluated by both the AI algorithm and expert radiologists for automated coronary calcium scoring and lung nodule detection are reported in Additional file 1: Table S1
Summary
Artificial intelligence (AI) in diagnostic radiology is undergoing rapid development. We present findings on the performance of an AI convolutional neural network (CNN) prototype (AI-RAD Companion, Siemens Healthineers) that automatically detects pulmonary nodules and quantifies coronary artery calcium volume (CACV) on low-dose chest CT (LDCT), and compare results to expert radiologists. Patients receiving LDCT scans typically have major risk factors that predispose them to coronary artery disease and would be highly advantageous to concurrently screen for both lung cancer and assess coronary calcification burden, which is a well-known marker for subsequent major cardiovascular adverse events [8,9,10]. Developed artificial intelligence (AI) deep learning methods using convolutional neural networks (CNN) have been used for the detection of lung nodules, which has been shown to improve detection sensitivity and reduce reading times [15,16,17]. On non-contrast chest CT scans, can introduce large margins of error due to motion and calcium location miscategorization; newer techniques could compensate for these limitations
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call