Background: The success of cardiac auscultation varies widely among medical professionals, which can lead to missed treatments for structural heart disease. Applying machine learning to cardiac auscultation could address this problem, but despite recent interest, few algorithms have been brought to clinical practice. We evaluated a novel suite of Food and Drug Administration‐cleared algorithms trained via deep learning on >15 000 heart sound recordings. Methods and Results: We validated the algorithms on a data set of 2375 recordings from 615 unique subjects. This data set was collected in real clinical environments using commercially available digital stethoscopes, annotated by board‐certified cardiologists, and paired with echocardiograms as the gold standard. To model the algorithm in clinical practice, we compared its performance against 10 clinicians on a subset of the validation database. Our algorithm reliably detected structural murmurs with a sensitivity of 85.6% and specificity of 84.4%. When limiting the analysis to clearly audible murmurs in adults, performance improved to a sensitivity of 97.9% and specificity of 90.6%. The algorithm also reported timing within the cardiac cycle, differentiating between systolic and diastolic murmurs. Despite optimizing acoustics for the clinicians, the algorithm substantially outperformed the clinicians (average clinician accuracy, 77.9%; algorithm accuracy, 84.7%.) Conclusions: The algorithms accurately identified murmurs associated with structural heart disease. Our results illustrate a marked contrast between the consistency of the algorithm and the substantial interobserver variability of clinicians. Our results suggest that adopting machine learning algorithms into clinical practice could improve the detection of structural heart disease to facilitate patient care.

Vasovagal syncope (VVS) is a common cardiovascular dysautonomic disorder that significantly impacts health and quality of life (QoL). Yoga has been shown to have a positive influence on cardiovascular autonomics. This study assessed the effectiveness of yoga therapy on the recurrence of VVS and QoL. We randomized subjects with recurrent reflex VVS (>3 episodes in the past 1 year) and positive head-up tilt test to guideline-directed therapy (Group 1) or yoga therapy (Group 2). Patients in Group 1 were advised guideline-directed treatment and Group 2 was taught yoga by a certified instructor. The primary endpoint was VVS recurrences and QoL. Between June 2015 and February 2017, 97 highly symptomatic VVS patients were randomized (Group 1: 47 and Group 2: 50). The mean age was 33.1 ± 16.6 years, male:female of 40:57, symptom duration of 17.1 ± 20.7 months, with a mean of 6.4 ± 6.1 syncope episodes. Over a follow-up of 14.3 ± 2.1 months Group 2 had significantly lower syncope burden compared with Group 1 at 3 (0.8 ± 0.9 vs. 1.8 ± 1.4, P < 0.001), 6 (1.0 ± 1.2 vs. 3.4 ± 3.0, P < 0.001), and at 12 months (1.1 ± 0.8 vs. 3.8 ± 3.2, P < 0.001). The Syncope functional score questionnaire was significantly lower in Group 2 compared with Group 1 at 3 (31.4 ± 7.2 vs. 64.1 ± 11.5, P < 0.001), 6 (26.4 ± 6.3 vs. 61.4 ± 10.7, P < 0.001), and 12 months (22.2 ± 4.7 vs. 68.3 ± 11.4, P < 0.001). For patients with recurrent VVS, guided yoga therapy is superior to conventional therapy in reducing symptom burden and improving QoL. Mechanism of syncope and yoga postures. The figure’s central portion shows the posture of dhyana (meditation) and the mechanism of syncope. The 10 asanas used in the study are depicted around the periphery. Counter-clockwise the from 12 o’clock position are Vrikshasana (tree pose), Ardha Kati Chakrasana (half waist wheel pose), Trikonasana (triangle pose), Pada Hastasana (hand under foot pose), Ardha Chakrasana (half wheel pose), Paschimottanasnana (seated forward bend), Ardha Ustrasana (half camel pose), Vakrasana (half spinal twist pose), Vajrasana (diamond pose), Vajrasana (diamond pose), and Setubandhasana (bridge pose).

Background: Marfan syndrome (MFS) is an autosomal dominant connective tissue disorder affecting multiple systems, particularly the cardiovascular system. The leading causes of death in MFS are aortopathies and valvular disease. We wanted to identify the frequency of arrhythmia and postural orthostatic tachycardia syndrome, length of hospital stay, health care–associated costs (HAC), and in‐hospital mortality in patients with MFS. Methods and Results: The National Inpatient Sample database from 2005 to 2014 was queried using International Classification of Diseases, Ninth Revision, Clinical Modification ( ICD‐9‐CM ) codes for MFS and arrhythmias. Patients were classified into subgroups: supraventricular tachycardia, ventricular tachycardia (VT), atrial fibrillation, atrial flutter, and without any type of arrhythmia. Data about length of stay, HAC, and in‐hospital mortality were also abstracted from National Inpatient Sample database. Adjusted HAC was calculated as multiplying HAC and cost‐to‐charge ratio; 12 079 MFS hospitalizations were identified; 1893 patients (15.7%) had an arrhythmia; and 4.9% of the patients had postural orthostatic tachycardia syndrome. Median values of length of stay and adjusted HAC in VT group were the highest among the groups (VT: 6 days, $18 975.8; supraventricular tachycardia: 4 days, $11 906.6; atrial flutter: 4 days, $11 274.5; atrial fibrillation: 5 days, $10431.4; without any type of arrhythmia: 4 days, $8336.6; both P =0.0001). VT group had highest in‐patient mortality (VT: 5.3%, atrial fibrillation: 4.1%, without any type of arrhythmia: 2.1%, atrial flutter: 1.7%, supraventricular tachycardia: 0%; P <0.0001) even after adjustment for potential confounders (without any type of arrhythmia versus VT; odds ratio [95% CI]: 3.18 [1.62–6.24], P =0.001). Conclusions: Arrhythmias and postural orthostatic tachycardia syndrome in MFS were high and associated with increased length of stay, HAC, and in‐hospital mortality especially in patients with VT.