Amir A. Amini

Assistant Professor of Diagnostic Radiology and Electrical Engineering

Professional Bio

I received the BS degree in electrical engineering from the University of Massachusetts, Amherst in 1983 with honors where at 18 I was the youngest member of the graduating class, and the MS and PhD degrees in 1984 and 1990 in electrical engineering with concentration in systems from the University of Michigan, Ann Arbor. My general research interests are in image processing, computer vision, and medical imaging. My specific research interests are in analysis of non-rigid motion restricted to medical imagery. The general underlying principle in my work is to use physically valid assumptions to track non-rigid motion. I have also been active in the area of deformable models and snakes for a number of years. Current projects are: (1) NMR Tagging. MR tagging is a medical imaging modality which creates signal voids in soft tissue by perturbing the tissue magnetization with selective RF excitations. We have developed deformable spline grids for tracking the MR tag lines. Techniques are also developed for measuring dense volumetric motion and deformations by using novel spline interpolations, as well as by utilizing volumetric tensor-product B-spline solids. This project is funded by the Whitaker Biomedical Engineering Foundation. (2) Optical flow techniques for non-rigid motion analysis of incompressible flow. We have developed a motion constraint equation for X-ray imaging, a special case of which is shown to be the Horn and Schunck's optical flow constraint. A new scalar function formulation for optical flow automatically enforces the incompressibility condition on the velocity field. This project is funded by The National Science Foundation. (3) Identifying vascular features with orientation specific quadrature filters and B-spline snakes and tensor-product B-spline surfaces. A quadrature filtering scheme is used in order to create an energy field for optimization and location of vascular features.

Representative Publications

  1. A. Amini, R. Curwen, and J. Gore "Snakes and Splines for Tracking Non-Rigid Heart Motion," European Conference on Computer Vision, University of Cambridge, UK, April 1996.

  2. A. Amini, "A Scalar Function Formulation For Optical Flow," European Conference on Computer Vision, Stockholm, Sweden, May 1994.

  3. A. Amini, R. Curwen, R. T. Constable, and J. Gore, "MR Physics-Based Snake Tracking and Dense Deformations from Tagged MR Cardiac Images," AAAI Symposium on Applications of Computer Vision to Medical Image Processing, Stanford University, Stanford, California, March 1994.

  4. A. Amini et al., ``Energy-Minimizing Deformable Grids for Tracking Tagged MR Cardiac Images,'' Computers in Cardiology, Durham, North Carolina, October 1992.

  5. A. Amini and J. S. Duncan, "Bending and stretching models for lv wall motion analysis from curves and surfaces", Image and Vision Computing, July/August 1992, pp. 418-430.

  6. A. Amini, T. Weymouth, and R. Jain, "Using dynamic programming for solving variational problems in vision", IEEE Transactiions on Pattern Analysis and Machine Intelligence, Vol. 12, No. 9, September 1990, pp. 855-867.

Other Interests

I like to play soccer, basketball, and various racket sports.

amini@noodle.med.yale.edu