In this study, Zhang et al. explores the potential of an automated computer vision pipeline for successful interpretation of echocardiogram analysis (disease detection and cardiac structure and function interpretation). Convolution neural network (CNN) models were taught and tested for multiple functions with the support of a sample of 14,035 collected echocardiograms over a 10 year span. Those assignments include identification of cardiac structures from 23 viewpoints as well as segmentation of cardiac chambers from multiple (5) perspectives, allowing for quantification of chamber dimensions and ejection fraction measurements. Additional disease detection models were developed, specifically for identification of hypertrophic cardiomyopathy, cardiac amyloid, and pulmonary arterial hypertension. Results proved the success of the CNN models to accurately identify views of cardiac chambers, indicate obscured chambers, and sanction segmentation of individual chambers. Collected cardiac structure measurements from the automated system, when compared to manual measurements, were found to be comparable. Additionally, the CNN model was able to detect hypertrophic cardiomyopathy (C statistic of .93), cardiac amyloidosis (C statistic of .87), and pulmonary arterial hypertension (C statistic of .85) with significant levels of accuracy. Overall, this pipeline marks a baseline for further automated interpretation developments to successfully analyze echocardiograms and support serial patient monitoring.