Biomedical Engineering

Distinguished Seminar Series: Andrew Laine, Professor and Chair, Biomedical Engineering, Columbia University

Andrew Laine, Ph.D.
Percy K. and Vida L. W. Hudson Professor
Chair, Department of Biomedical Engineering
Professor of Radiology (Physics), Department of Radiology
Director, Heffner Biomedical Imaging Laboratory
Columbia University, New York, NY

“Advances in Quantitative Analysis of Medical Images: Functional Imaging and Modeling of the Heart, Brain and Vasculature”

Dynamic cardiac metrics, including strains and displacements, can provide a quantitative approach to evaluate cardiac function. However, in current clinical diagnosis, strain measures in 2D are used despite the fact that cardiac motions are complex changes in 4D. Recent advances in 4D ultrasound enable the capability to capture such complex motion in real time. In our previous work, a 4D optical flow based motion tracking algorithm was developed to extract full 4D dynamic cardiac metrics from such 4D ultrasound data. In order to quantitatively evaluate this method, coronary artery occlusion experiments at various locations were performed on five canine hearts with 4D ultrasound and sonomicrometry data acquired during the occlusion. Optical flow displacement was then mapped onto a finite element field fitted model. Corresponding 4D ultrasound data from these experiments were then analyzed. Estimated principal strains were directly compared to those recorded by sonomicrometry showing strong agreement. This was the first validation study of optical flow based strain estimation for 4D cardiac ultrasound including a direct comparison with sonomicrometry on in vivo data. Finally, a clinical study is presented to validate the performance of 4D cardiac ultrasound strain measures to cardiac MRI using 3D DENSE and 3D CSPAMM as gold standards of myocardial strain.

The talk will also include recent research on other methods to quantify dynamic metrics in medical imaging such as hyper-spectral imaging of the retina for classification of dusen related to AMD, biomedical informatics and the integration of multimodal longitudinal data for diagnosis and treatment of chronic diseases.

When: Thursday, May 16, 2013 4:10 PM

Where: 1005 GBSF