Analytics Pipeline for Left and Right Ventricle Segmentation on Cardiac MRI using Deep Learning. In Data Science & Engineering Master of Advanced Study (DSE MAS) Capstone Projects

This is the capstone project for MAS DSE program. The goal of the capstone project is to complete an end to end analysis of a large dataset with big data characteristics. The​ ​goal​ ​of​ ​this​ ​project​ ​is​ ​to​ use magnetic resonance imaging(MRI) data to provide an end-to-end analytics pipeline for left and right ventricle (LV & RV) segmentation. For the left ventricle, we map both the epicardium and endocardium surfaces. The endocardium segment is the area of the left ventricle contained by the inside of the wall of the left ventricle, and is the most reliable way to calculate the ejection fraction of the heart. The left ventricle epicardium segment is the area contained by the outside of the wall of the left ventricle, and once the contours of the inside and outside of the wall are known, the contours of the myocardium, the actual wall, can be calculated. The contours of the myocardium are used in detecting the severity of damage done by heart attacks. For the right ventricle, we segment the endocardium. Since the process of analyzing cardiac images is time consuming and labor intensive, our project aims to apply deep learning to this process to achieve consistent segmentation results efficiently. We also compare the performance of 2D UNET, 3D UNET and Densenet models across various datasets. The datasets used in this project are Sunnybrook Cardiac dataset, Automated Cardiac Diagnosis Challenge dataset, MICCAI 2012 Right Ventricle Segmentation challenge dataset and CAP 2011 Left Ventricle Segmentation challenge dataset.