Massive Parallelization of the Conjugated Gradients Algorithm applied to the reconstruction of Electrical Impedance Tomography Images

• Grant number: 11/01194-8
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: April 01, 2011
• End date: March 31, 2012
• Field of knowledge: Engineering - Biomedical Engineering - Medical Engineering
• Principal Investigator: Thiago de Castro Martins
• Grantee: Jacqueline de Miranda Kian
• Host Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

The reconstruction of Electrical Impedance Tomography Images can be approached as an optimization problem, where one minimizes the differences between observed data and a syntetic model, simulated using the Finite Element Method (FEM). Under this approach, the bulk of computational costs is in spent solving the linear systems produced by the FEM. This project's goal is to research techniques for massive parallelization of the Conjugated Gradients Algorithm to solve such systems, considering the specifics of the studied problem.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

MARTINS, THIAGO DE CASTRO; KIAN, JACQUELINE DE MIRANDA; SATO, ANDRE KUBAGAWA; GUERRA TSUZUKI, MARCOS DE SALES; IEEE. Matrix-Vector Multiplication and Triangular Linear Solver Using GPGPU for Symmetric Positive Definite Matrices Derived from Elliptic Equations. 6TH INTERNATIONAL CONFERENCE ON SOFT COMPUTING AND INTELLIGENT SYSTEMS, AND THE 13TH INTERNATIONAL SYMPOSIUM ON ADVANCED INTELLIGENT SYSTEMS, v. N/A, p. 6-pg., 2012-01-01. (11/01194-8, 10/19646-0, 09/07173-2)