ML Segmentation on Cardiac MR Images
Authors: Daniel Chang, Kimi Lee
Faculty Supervisor: Ilmi Yoon
Department: Computer Science
Cardiac MR Imaging presents a dataset for diagnosing cardiovascular diseases, yet manual segmentation remains a bottleneck due to its labor-intensive nature. This study employs the ACDC dataset to automate cardiac MRI segmentation and subsequent classification using computer science methodologies. Leveraging image processing techniques and machine learning algorithms, we delineate key cardiac structures, such as the left ventricle, right ventricle, myocardium, and background, from MRI images. Our approach involves feature extraction, pixel clustering, and machine learning model training to achieve accurate segmentation and classification results. By integrating computer science principles, including data-driven methods and algorithmic optimization, we aim to streamline cardiac MRI analysis, enhancing efficiency and reliability in clinical settings. This research contributes to advancing computer-assisted diagnostic tools for cardiovascular health assessment, offering promising avenues for improved patient care and medical decision-making.