Video based deep learning evaluation of cardiac ultrasound accurately identifies cardiomyopathy and predict ejection fraction, the most common metric of cardiac function.• Using human tracings obtained during standard clinical workflow, deep learning semantic segmentation accurately labels the left ventricle frame-by-frame, including in frames without prior human annotation.• Computational cardiac function analysis allows for beat-by-beat assessment of ejection fraction, which more accurately assesses cardiac function in patients with atrial fibrillation, arrhythmias, and heart rate variability.
Abstract-A group of relatively uncommon but important genetic cardiovascular diseases (GCVDs) are associated with increased risk for sudden cardiac death during exercise, including hypertrophic cardiomyopathy, long-QT syndrome, Marfan syndrome, and arrhythmogenic right ventricular cardiomyopathy. These conditions, characterized by diverse phenotypic expression and genetic substrates, account for a substantial proportion of unexpected and usually arrhythmia-based fatal events during adolescence and young adulthood. Guidelines are in place governing eligibility and disqualification criteria for competitive athletes with these GCVDs (eg, Bethesda Conference No. 26 and its update as Bethesda Conference No. 36 in 2005). However, similar systematic recommendations for the much larger population of patients with GCVD who are not trained athletes, but nevertheless wish to participate in any of a variety of recreational physical activities and sports, have not been available. The practicing clinician is frequently confronted with the dilemma of designing noncompetitive exercise programs for athletes with GCVD after disqualification from competition, as well as for those patients with such conditions who do not aspire to organized sports. Indeed, many asymptomatic (or mildly symptomatic) patients with GCVD desire a physically active lifestyle with participation in recreational and leisure-time activities to take advantage of the many documented benefits of exercise. However, to date, no reference document has been available for ascertaining which types of physical activity could be regarded as either prudent or inadvisable in these subgroups of patients. Therefore, given this clear and present need, this American Heart Association consensus document was constituted, based largely on the experience and insights of the expert panel, to offer recommendations governing recreational exercise for patients with known GCVDs.
Echocardiography uses ultrasound technology to capture high temporal and spatial resolution images of the heart and surrounding structures, and is the most common imaging modality in cardiovascular medicine. Using convolutional neural networks on a large new dataset, we show that deep learning applied to echocardiography can identify local cardiac structures, estimate cardiac function, and predict systemic phenotypes that modify cardiovascular risk but not readily identifiable to human interpretation. Our deep learning model, EchoNet, accurately identified the presence of pacemaker leads (AUC = 0.89), enlarged left atrium (AUC = 0.86), left ventricular hypertrophy (AUC = 0.75), left ventricular end systolic and diastolic volumes (R 2 = 0.74 and R 2 = 0.70), and ejection fraction (R 2 = 0.50), as well as predicted systemic phenotypes of age (R 2 = 0.46), sex (AUC = 0.88), weight (R 2 = 0.56), and height (R 2 = 0.33). Interpretation analysis validates that EchoNet shows appropriate attention to key cardiac structures when performing human-explainable tasks and highlights hypothesis-generating regions of interest when predicting systemic phenotypes difficult for human interpretation. Machine learning on echocardiography images can streamline repetitive tasks in the clinical workflow, provide preliminary interpretation in areas with insufficient qualified cardiologists, and predict phenotypes challenging for human evaluation.
Distinct patterns of aortic dilatation in patients with bicuspid aortic valves call for an individualized degree of aortic replacement to minimize late aortic complications and reoperation. Patients in clusters III and IV should have transverse arch replacement (plus concomitant root replacement in cluster IV). Patients in cluster I should undergo complete aortic root replacement, whereas in patients in cluster II supracommissural ascending aortic grafting is adequate.
The Loeys–Dietz syndrome (LDS) is a connective tissue disorder affecting the cardiovascular, skeletal, and ocular system. Most typically, LDS patients present with aortic aneurysms and arterial tortuosity, hypertelorism, and bifid/broad uvula or cleft palate. Initially, mutations in transforming growth factor‐β (TGF‐β) receptors (TGFBR1 and TGFBR2) were described to cause LDS, hereby leading to impaired TGF‐β signaling. More recently, TGF‐β ligands, TGFB2 and TGFB3, as well as intracellular downstream effectors of the TGF‐β pathway, SMAD2 and SMAD3, were shown to be involved in LDS. This emphasizes the role of disturbed TGF‐β signaling in LDS pathogenesis. Since most literature so far has focused on TGFBR1/2, we provide a comprehensive review on the known and some novel TGFB2/3 and SMAD2/3 mutations. For TGFB2 and SMAD3, the clinical manifestations, both of the patients previously described in the literature and our newly reported patients, are summarized in detail. This clearly indicates that LDS concerns a disorder with a broad phenotypical spectrum that is still emerging as more patients will be identified. All mutations described here are present in the corresponding Leiden Open Variant Database.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.