Matthias Chung, PhD

Associate Professor, Ecas, Mathematics

John Oshinski, PhD

Professor, Emory School Of Medicine And Georgia Tech School Of Engineering, Biomedical Engineering

Accelerated Reconstruction of 5D Whole-Heart Cardiac MRI

Whole-heart Magnetic Resonance Imaging (MRI) is an invaluable noninvasive diagnostic tool for evaluating cardiac anatomy and function, and for monitoring the progression of cardiac diseases. The acquired raw MRI data needs to undergo a computationally expensive and ill-posed inversion process to obtain meaningful and detailed time-resolved 3D images of the heart. The respiratory and contractile movement of the heart causes artifacts that pose additional challenges to acquisition and reconstruction. Current state-of-the-art inversion techniques require more than 10 hours to reconstruct cardiac and respiratory-resolved images, rendering the technique untenable for clinical use. Our aim is to make the inversion process viable for routine clinical use. We will utilize a recently developed novel method based on the variable projection framework. In a preliminary study, we could reduce computation times significantly. In this proposed project, we will develop a carefully curated and integrated cardiac MRI inversion method, develop computational phantoms for calibration, integrate a mathematical model to compensate for cardiac and respiratory motion, and test the method on an extensive set of clinical image data. Our research has the potential to provide a fundamental shift in the use of MRI for diagnosis and intervention in cardiovascular imaging.

Hans E Grossniklaus, MD, MBA

Professor, Emory School Of Medicine, Ophthalmology

Ahmet Coskun, PhD

Assistant Professor, Emory University School Of Medicine and Georgia Institute Of Technology, Wallace H. Coulter Department Of Biomedical Engineering

Vasculogenic Mechanisms in Uveal Melanoma

Uveal melanoma (UM), a lethal tumor, is the most common primary ocular tumor. Mutations in UM cells that result in aggressive tumor behavior are not enough to promote tumor growth and metastasis and architectural tumor features are needed for UM progression. The most important architectural feature is UM vascularity. Vasculogenesis, the process that results in UM vascularity, will be studied in this proposal. UM vasculogenesis appears to occur from three distinct, progressive mechanisms- (1) macrophage infiltration with channel formation; (2) lining of these channels with endothelial cells and UM cells (vasculogenic mimicry, or VM), and (3) development of complex fibrovascular septae within the UM. These mechanisms are hypoxia/HIF1α dependent-macrophage infiltration/channel formation from hypoxia induced macrophage production MMP2/9; hypoxia induced VEGF production from M2 macrophages resulting in VM; hypoxia induced P4HA1/2 collagen production from UM cells and fibroblasts that result in complex fibrovascular patterns. These processes will be visualized in human UM tissue using spatiomic methods and will be reproduced in mouse models of UM.

Kimberly Hoang, MD

Assistant Professor, School Of Medicine, Neurosurgery

Francisco Robles, PhD

Assistant Professor, Emory School Of Medicine And Georgia Tech School Of Engineering, Biomedical Engineering

Towards In-Vivo, Intraoperative Image Guided Brain Tumor Margin Assessment with Quantitative Oblique Back Illumination Microscopy

During the surgical removal of tumors of the brain, the surgeon’s goal is to remove as much of the tumor as is safely possible and still preserve normal brain. In tumors of brain itself, the edge of the tumor is quite indistinct, gradually blending into the normal brain. To address this, neurosurgeons use the operating microscope, fluorescent dyes, and navigation with imaging tools such as CT scans and MRI scans. Though these are powerful techniques, they do not provide details of the edge of the tumor removal down to a cellular level. The investigative proposal will develop the use of a handheld microscope that is small enough that it can be held along the edge of the tumor removal margin. The technique is call quantitative oblique back illumination microscopy, or qOBM for short. This will be done in subjects who are requiring surgery and we will determine the safety of its use and the ability of this new tool to assess the edge of the tumor removal in real time as the surgeon is working. The information from this initial study in patients will be used to create a more extensive proposal seeking larger federal and foundation grant funding. Ideally, the use of this new device will allow greater extent of tumor removal and a more accurate detection of brain that needs to be preserved.