BME PhD Preliminary Exam Announcement for Ruhi Sharmin (P. Vlachos, advisor) Everyone is invited to attend the public presentation beginning at 9:00am. Title: Signal Processing and Machine Learning For Cardiac Diagnostics: ECG, Echocardiography, and MRI Applications Date: October 23, 2024 Time: 9:00am Location: LPRC MRGN 229 Committee: Dr. Pavlos P. Vlachos, Chair Members - Dr. Edward J. Delp, Dr. Craig J. Goergen, Dr. Ilias Bilionis, Dr. Brett A. Meyers Abstract: Cardiovascular diagnostics are crucial for effective treatment and management of critical cardiac conditions, yet current methods often lack the accuracy and clinical relevance needed to make timely and informed decisions. Despite advancements in computational techniques, gaps remain in utilizing multimodal data from electrocardiograms (ECGs), echocardiograms, and magnetic resonance imaging (MRI) for improving diagnostics. This study aims to address this gap by developing novel methodologies that integrate machine learning, signal and image processing, and biomechanical analysis. Our research focuses on four key areas. First, we aim to improve the detection of Atrial Fibrillation (AFib), a common heart rhythm disorder, by developing an interpretable and efficient feature extraction method from ECG data (Aim 1). By applying advanced algorithms like XGBoost (eXtreme Gradient Boosting), we expect to enhance AFib detection in real-time, even in shorter ECG recordings. Second, we are developing a hybrid framework to automate the segmentation of cardiac structures, such as the left ventricle (LV), from echocardiograms (Aim 2). This approach combines traditional image processing with deep learning, improving the accuracy and consistency of clinical measurements like ejection fraction. Third, we propose a novel fusion method that integrates Color Doppler echocardiography with 4D Flow MRI (Aim 3). This technique will enhance the accuracy of blood flow velocity measurements, providing a more complete understanding of cardiovascular flow dynamics—especially important in complex heart conditions. Lastly, we will investigate the biomechanical alterations in the Systemic Right Ventricle (SRV) during the critical transition from fetal to neonatal stages (Aim 4). By introducing advanced hemodynamic markers such as energy loss and vortex strength, this study seeks to offer new insights into congenital heart defects, particularly Hypoplastic Left Heart Syndrome (HLHS). Combinedly, these aims will significantly advance cardiovascular diagnostics by developing clinically interpretable, automated tools for early and accurate detection of cardiac abnormalities. The utilization of important information from ECG, echocardiography, and MRI will provide more reliable diagnostic capabilities, ultimately improving patient outcomes and informing clinical decision-making in a wide range of cardiovascular conditions. -- Bmeroundtable-list mailing list Bmeroundtable-list@ecn.purdue.edu https://engineering.purdue.edu/ECN/mailman/listinfo/bmeroundtable-list