Our approach segments the left ventricle (LV) cavity in each short-axis (SAX) slice, computes the LV area and volume for each slice, then sums all per-slice LV volumes to estimate the overall LV volume. We use a U-Net, a deep learning model originally created for image segmentation in biomedical applications but have been applied to other domains as well. In our approach, a U-Net is trained on the segmentation task by predicting the LV contour in each input cardiac magnetic resonance (CMR) image. The contours are then used as input to a separate process to calculate the LV volume. In this volume calculation process, end-systole (ES) and end-diastole (ED) frames are determined in each slice, then summed across all slices to determine the LV volume for each patient. We make use of three publicly available data sources: Sunnybrook Cardiac Data (SCD), Automated Cardiac Diagnosis Challenge (ACDC), and Kaggle Second Annual Data Science Bowl. We discuss findings from our investigation into different techniques for processing and analyzing CMR images and present the methods giving best performance in an end-to-end analytics pipeline for LV segmentation and volume estimation. This pipeline can serve as an initial step towards analyzing CMR at scale to aid in non-invasive cardiac disease detection.