Einstein, Andrew Jeffrey

Abstract Background and Aims Classification and risk stratification in aortic (AR), mitral (MR), and tricuspid regurgitation (TR) remains a significant clinical challenge. This study aimed to develop an artificial intelligence (AI) system to assess valvular regurgitation and stratify MR-progression risk. Methods Using transthoracic echocardiograms (TTEs) at two sites (internal development/test, external test), the DELINEATE-Regurgitation system was developed to classify AR, MR, and TR severity using colour Doppler videos. Methods of summating video-level classifications into study-level predictions were tested, comparing single-view with multiview approaches integrating predictions across multiple videos. Model agreement with cardiologists was assessed by weighted kappa. A separate AI system (DELINEATE-MR-Progression) analysing colour Doppler videos was developed to predict which patients with mild, mild–moderate, and moderate MR were most likely to progress to moderate–severe or severe MR with analysis by Kaplan–Meier and Cox proportional hazards models. Results A total of 71 660 TTEs with 1 203 980 colour Doppler videos were included. The weighted kappa in internal/external test sets for regurgitation classification was 0.81/0.76 for AR, 0.76/0.72 for MR, and 0.73/0.64 for TR using a multiview approach taking all colour Doppler videos in a study, demonstrating substantial agreement with cardiologist interpretation with superiority of multiview over single view approaches. In the progression analysis, the AI score stratified MR-progression risk even when controlled for clinical factors known to be associated with MR progression [internal test set hazard ratio 4.1 (95% confidence interval 2.5–6.6)]. Conclusions An AI system can accurately classify AR, MR, and TR and predict MR progression beyond currently known risk factors.

ABSTRACTObjectivesIdentifying the distributions and determinants of fluoroscopy time for invasive coronary angiography (ICA) and percutaneous coronary intervention (PCI).BackgroundICA and PCI are significant contributors to radiation exposure from medical imaging in the US. Fluoroscopy time is a potentially modifiable determinant of radiation exposure for these procedures, but has not been well characterized in contemporary practice.MethodsWe evaluated the distribution of fluoroscopy time in patients undergoing ICA and/or PCI in the CathPCI Registry®, stratifying patients by numerous clinical scenarios. Hierarchical models were used to determine patient, procedure, operator and hospital‐level factors associated with fluoroscopy time for these procedures.ResultsOur study included a total of 3,295,348 ICA and PCI procedures performed by 9,600 operators from January 2005 through June 2009. There was wide variation in fluoroscopy times for these procedures with median [IQR] fluoroscopy times of 2.6 [1.7–4.5] minutes for ICA, 6.7 [4.2–10.8] minutes for ICA in patients with prior coronary artery bypass grafting (CABG), 10.1 [6.0–17.4] minutes for PCI, 10.7 [7.0–16.9] minutes for PCI with ICA, and 16.0 [10.6–24.0] minutes for PCI and ICA in patients with prior CABG. Prolonged fluoroscopy times (>30 minutes) were rare for ICA, but occurred in 6.7% of PCIs and 14.7% of PCIs in patients with prior CABG. After accounting for patient characteristics and procedure complexity, operator and hospital‐level factors explained nearly 20% of the variation in fluoroscopy time.ConclusionsFluoroscopy times vary widely during ICA and PCI with operator and hospital‐level factors contributing substantially to these differences. A better understanding of potentially modifiable sources of this variation will elucidate opportunities for enhancing the radiation safety of these procedures. © 2013 Wiley Periodicals, Inc.