We discuss the scalable parallel solution of the Poisson equation on irregularly shaped domains discretized by finite differences. The symmetric positive definite system is solved by the preconditioned conjugate gradient algorithm with smoothed aggregation (SA) based algebraic multigrid (AMG) preconditioning. We investigate variants of the implementation of SA-AMG that lead to considerable impr...

Automatic differentiation has been involved for long in applied mathematics as an alternative to finite difference to improve the accuracy of numerical computation of derivatives. Each time a numerical minimization is involved, automatic differentiation can be used. In between formal derivation and standard numerical schemes, this approach is based on software solutions applying mechanically th...

Objective: Remote Ischemic Preconditioning (RIPC) as the transient ischemia and reperfusion of the arm is a promising method for protecting different tissue from future ischemia. These effects might be mediated through vascular and endothelial growth factor (VEGF) pathway. We investigated the influence of RIPC on diabetic macular edema (DME) as a chronic ischemic condition in patients who were ...

The 3D finite difference modeling of the wave propagation of M>8 earthquakes in subduction zones in a realistic-size earth is very computationally intensive task. We use a parallel finite difference code that uses second order operators in time and fourth order differences in space on a staggered grid. We develop an efficient parallel program using message passing interface (MPI) and a kinemati...

A finite-difference Micromagnetic simulation code in MATLAB is presented with Graphics Processing Unit (GPU) acceleration. The high performance of Graphics Processing Unit (GPU) is demonstrated compared to a typical Central Processing Unit (CPU) based code. The speed-up of GPU to CPU is shown to be greater than 30 for problems with larger sizes on a mid-end GPU in single precision. The code is ...

We consider the resolution of an inverse problem for the identification of some reduced nonlinear diffusive models. The modal identification method is dealt with along with a gradient-type optimization method. Two distinct approaches are considered and compared. On one hand the classical finite differences approach and, on the other hand, the adjoint method for the computation of the objective ...

(p<0.005) and (B) BOLD changes (p<0.005) during right hand movement displayed on an inflated cortical surface with white arrow indicating the central sulcus. (C) shows the locations of peak BOLD changes and time course of BOLD percentage change derived from these two regions (baseline is from -4 s to 0 s). ACCOMPANYING EVENT-RELATED DECREASE OF ALPHA BAND EEG AND SUSTAINED NEGATIVE BOLD RESPONS...