Citations

Abstract Purpose Echocardiography (echo) is widely used for right ventricular (RV) assessment. Current techniques for RV evaluation require additional imaging and manual analysis; machine learning (ML) approaches have the potential to provide efficient, fully automated quantification of RV function. Methods An automated ML model was developed to track the tricuspid annulus on echo using a convolutional neural network approach. The model was trained using 7791 image frames, and automated linear and circumferential indices quantifying annular displacement were generated. Automated indices were compared to an independent reference of cardiac magnetic resonance (CMR) defined RV dysfunction (RVEF < 50%). Results A total of 101 patients prospectively underwent echo and CMR: Fully automated annular tracking was uniformly successful; analyses entailed minimal processing time (<1 second for all) and no user editing. Findings demonstrate all automated annular shortening indices to be lower among patients with CMR‐quantified RV dysfunction (all P < .001). Magnitude of ML annular displacement decreased stepwise in relation to population‐based tertiles of TAPSE, with similar results when ML analyses were localized to the septal or lateral annulus (all P ≤ .001). Automated segmentation techniques provided good diagnostic performance (AUC 0.69–0.73) in relation to CMR reference and compared to conventional RV indices (TAPSE and S ′) with high negative predictive value (NPV 84%–87% vs 83%–88%). Reproducibility was higher for ML algorithm as compared to manual segmentation with zero inter‐ and intra‐observer variability and ICC 1.0 (manual ICC: 0.87–0.91). Conclusions This study provides an initial validation of a deep learning system for RV assessment using automated tracking of the tricuspid annulus.

Background— The appropriate indication for coronary computed tomographic angiography (CTA) as a part of preoperative evaluation has not been defined yet. We investigated the value of coronary CTA in patients undergoing noncardiac surgery. Methods and Results— We included 844 patients (median age, 67 years; male sex, 62%) who underwent coronary CTA for screening of coronary artery disease before noncardiac surgery. Clinically determined revised cardiac risk index were compared with the extent and severity of coronary artery disease assessed by coronary CTA. Perioperative major cardiac event (PMCE), defined as cardiac death, myocardial infarction, or pulmonary edema within postoperative 30 days, developed in 25 patients (3.0%). Significant coronary CTA finding was defined as >3 any lesions with ≥1 (diameter stenosis ≥70%) stenosis based on the relationship between the severity of coronary artery disease and PMCE risk. The risk of PMCE was 14.0% in patients with significant CTA findings, whereas 2.2% of patients without significant CTA findings regardless of revised cardiac risk index score. The predictive performance of revised cardiac risk index could be improved significantly after addition of significant coronary CTA findings (c-statistics=0.631 versus 0.757; net reclassification improvement=0.923; integrated discrimination improvement=0.051). On the basis of revised cardiac risk index and coronary CTA, the risk of PMCE could be estimated with sensitivity, specificity, positive predictive value, and negative predictive value of 76%, 73%, 8%, and 99%, respectively. Conclusions— Addition of coronary CTA to clinical risk improved perioperative risk stratification. Absence of significant coronary CTA findings conferred low PMCE risk with high specificity and negative predictive value regardless of clinical risk. Coronary CTA may improve perioperative risk stratification in patients undergoing noncardiac surgery.