In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
will defend his proposal
A Novel Deep Learning Approach for LV MRI Analysis in Short-Axis MRI
AbstractPhysicians use Magnetic Resonance Imaging (MRI) scans to obtain relevant images of the cardiac areas to assess the structural and functional features for cardiovascular diagnosis and disease management in a non-invasive manner. Currently, it takes physicians several minutes to diagnose a patient's condition and the obtained results are not easily reproducible. Using an improved automatic process to determine heart parameters and function can lead to a quicker, coherent diagnosis and generate a repeatable diagnostic process. We introduce a novel deep learning architecture called BNU-Net for the purpose of cardiac segmentation based on short-axis MRI images. Its name is derived from the Batch Normalized (BN) U-Net architecture for medical image segmentation. New generations of deep neural networks (NN) are called convolutional NN (CNN). CNNs are supervised training models that are trained to learn hierarchies of features automatically and robustly perform classification. Our architecture consists of an encoding path for feature extraction and a decoding path that enables precise localization. We compare this approach with a parallel approach named U-Net. The presented work (i) facilitates various image preprocessing techniques, which includes affine transformations and elastic deformations, and (ii) segments the preprocessed images using the new deep learning architecture. The experimental results reveal that our approach accomplishes comparable or better performance than other state-of-the-art approaches in terms of the Dice coefficient and the average perpendicular distance.
Date: Monday, November 18, 2019
Time: 11:00 AM - 12:00 PM
Place: PGH 501D
Advisor: Dr. Nikolaos V. Tsekos
Faculty, students, and the general public are invited.