Thomas A. Goldstein

Learn More
Computing with large kernel or similarity matrices is essential to many state-of-the-art machine learning techniques in classification, clustering, and dimension-ality reduction. The cost of forming and factoring these kernel matrices can become intractable for large datasets. We introduce an an adaptive column sampling technique called Accelerated(More)
—Sparse approximations using highly over-complete dictionaries is a state-of-the-art tool for many imaging applications including denoising, super-resolution, compressive sensing, light-field analysis, and object recognition. Unfortunately, the applicability of such methods is severely hampered by the computational burden of sparse approximation: these(More)
OBJECTIVES To quantify resting myocardial blood flow (MBF) in the left ventricular (LV) wall of HCM patients and to determine the relationship to important parameters of disease: LV wall thickness, late gadolinium enhancement (LGE), T2-signal abnormalities (dark and bright signal), LV outflow tract obstruction and age. MATERIALS AND METHODS Seventy(More)
PURPOSE Myocardial blood volume (MBV) may provide complementary information about myocardial oxygen needs and viability. The aim of this study is to examine a Cardiovascular Magnetic Resonance (CMR) perfusion method to quantify the changes in MBV, in comparison with the radiolabeled 99mTc-Red-Blood-Cell (RBC) method. METHODS Normal mongrel dogs (n=12)(More)
  • 1