Abstract: Echocardiography deep learning and cardiovascular outcomes are described. An echocardiogram analysis module may include a deep learning model to generate a video output for an input echocardiogram video, the deep learning model comprising a convolutional neural network and at least one dense layer. The echocardiogram analysis module may further include a cardiac prediction generator to generate a cardiac prediction based on video outputs generated for a plurality of input echocardiogram videos of an echocardiogram study, the cardiac prediction comprising a measurement prediction or a classification prediction.

Abstract: Echocardiography deep learning and cardiovascular outcomes are described. An echocardiogram analysis module may include a deep learning model to generate a video output for an input echocardiogram video, the deep learning model comprising a convolutional neural network and at least one dense layer. The echocardiogram analysis module may further include a cardiac prediction generator to generate a cardiac prediction based on video outputs generated for a plurality of input echocardiogram videos of an echocardiogram study, the cardiac prediction comprising a measurement prediction or a classification prediction.

Abstract: The disclosure provides methods of assessing a subject's risk of ascending thoracic aortic aneurysm or descending thoracic aortic aneurysm by detecting one or more single nucleotide polymorphisms (SNP) in one or more specific genes. The disclosure also provides methods of calculating a polygenic score to assess a subjects risk. Such methods may be used, for example, to identify asymptomatic subjects at risk for aneurysm.

Abstract: Provided herein are methods for diagnosing cardiomyopathy by evaluating a sample from a subject for the presence of a population of activated fibroblasts comprising a particular genetic signature (e.g., a genetic signature comprised of differential expression of one or more gene products relative to a normal subject or a population of normal subjects). The one or more gene products may include POSTN, COL22A1, and/or THBS4. Also provided herein are methods for treating cardiomyopathy and for modulating the activity of one or more gene products associated with activation of cardiac fibroblasts in a subject using an agent. Compositions comprising an agent for treating cardiomyopathy and kits for diagnosing cardiomyopathy are also provided by the present disclosure.

Abstract: Methods for detecting disorders associated with atrial fibrillation, and, thus, at risk for stroke and/or heart failure) in a subject based on a monitoring plasma levels of apelin are provided. Diagnostic compositions for the detection of such disorders are additionally provided.