Deep Learning Algorithm for Automated Cardiac Murmur Detection via a Digital Stethoscope Platform.

Background Clinicians vary markedly in their ability to detect murmurs during cardiac auscultation and identify the underlying pathological features. Deep learning approaches have shown promise in medicine by transforming collected data into clinically significant information. The objective of this research is to assess the performance of a deep learning algorithm to detect murmurs and clinically significant valvular heart disease using recordings from a commercial digital stethoscope platform. Methods and Results Using >34 hours of previously acquired and annotated heart sound recordings, we trained a deep neural network to detect murmurs. To test the algorithm, we enrolled 962 patients in a clinical study and collected recordings at the 4 primary auscultation locations. Ground truth was established using patient echocardiograms and annotations by 3 expert cardiologists. Algorithm performance for detecting murmurs has sensitivity and specificity of 76.3% and 91.4%, respectively. By omitting softer murmurs, those with grade 1 intensity, sensitivity increased to 90.0%. Application of the algorithm at the appropriate anatomic auscultation location detected moderate-to-severe or greater aortic stenosis, with sensitivity of 93.2% and specificity of 86.0%, and moderate-to-severe or greater mitral regurgitation, with sensitivity of 66.2% and specificity of 94.6%. Conclusions The deep learning algorithm's ability to detect murmurs and clinically significant aortic stenosis and mitral regurgitation is comparable to expert cardiologists based on the annotated subset of our database. The findings suggest that such algorithms would have utility as front-line clinical support tools to aid clinicians in screening for cardiac murmurs caused by valvular heart disease. Registration URL: https://clinicaltrials.gov; Unique Identifier: NCT03458806.

Urinary Metal Levels after Repeated Edetate Disodium Infusions: Preliminary Findings.

Environmentally acquired lead and cadmium are associated with increased cardiovascular disease risk. In the Trial to Assess Chelation Therapy, up to 40 infusions with edetate disodium over an approximately one-year period lowered the cardiovascular disease risk in patients with a prior myocardial infarction. We assessed whether a reduction in surrogate measures of total body lead and cadmium, post-edetate disodium urine lead and pre-edetate urine cadmium, could be detected after repeated edetate disodium-based infusions compared to the baseline. Fourteen patients with coronary artery disease received multiple open-label edetate disodium infusions. The urine metals pre- and post-edetate infusion, normalized for urine creatinine, were compared to urine levels pre and post final infusion by a paired t-test. Compared with the pre-edetate values, post-edetate urine lead and cadmium increased by 3581% and 802%, respectively, after the first infusion. Compared to baseline, post-edetate lead decreased by 36% (p = 0.0004). A reduction in post-edetate urine lead was observed in 84% of the patients after the final infusion. Pre-edetate lead decreased by 60% (p = 0.003). Pre-edetate lead excretion became undetectable in nearly 40% of patients. This study suggests that edetate disodium-based infusions may decrease the total body burden of lead. However, our data suggest no significant reduction in the body burden of cadmium.