Accelerating image registration of MRI by GPU-based parallel computation

Automatic image registration for MRI applications generally requires many iteration loops and is, therefore, a time-consuming task. This drawback prolongs data analysis and delays the workflow of clinical routines. Recent advances in the massively parallel computation of graphic processing units (GPUs) may be a solution to this problem. This study proposes a method to accelerate registration calculations, especially for the popular statistical parametric mapping (SPM) system. This study reimplemented the image registration of SPM system to achieve an approximately 14-fold increase in speed in registering single-modality intrasubject data sets. The proposed program is fully compatible with SPM, allowing the user to simply replace the original image registration library of SPM to gain the benefit of the computation power provided by commodity graphic processors. In conclusion, the GPU computation method is a practical way to accelerate automatic image registration. This technology promises a broader scope of application in the field of image registration.     

Minimizing the total weighted late work in scheduling of identical parallel processors with communication delays

This paper addresses a certain type of scheduling problem that arises when a parallel computation is to be executed on a set of identical parallel processors. It is assumed that if two precedence-related tasks are processed on two different processors, due to the information transferring, there will be a task-dependent communication delay between them. For each task, a processing time, a due date and a weight is given while the goal is to minimize the total weighted late work. An integer linear mathematical programming model and a branch-and-bound algorithm have been developed for the proposed problem. Comparing the results obtained by the proposed branch-and-bound algorithm with those obtained by CPLEX, indicates the effectiveness of the method.     

Optimal broadcast on parallel locality models

In this paper matching upper and lower bounds for broadcast on general purpose parallel computation models that exploit network locality are proven. These models try to capture both the general purpose properties of models like the PRAM or BSP on the one hand, and to exploit network locality of special purpose models like meshes, hypercubes, etc., on the other hand. They do so by charging a cost l(|i−j|) for a communication between processors i and j, where l is a suitably chosen latency function.An upper bound T(p)=∑_i=0^loglogp2ⁱ·l(p^1/2i) on the runtime of a broadcast on a p processor H-PRAM is given, for an arbitrary latency function l(k).The main contribution of the paper is a matching lower bound, holding for all latency functions in the range from l(k)=Ω(logk/loglogk) to l(k)=O(log²k). This is not a severe restriction since for latency functions l(k)=O(logk/log¹⁺log(k)) with arbitrary >0, the runtime of the algorithm matches the trivial lower bound Ω(logp) and for l(k)=Θ(log¹⁺k) or l(k)=Θ(k), the runtime matches the other trivial lower bound Ω(l(p)). Both upper and lower bounds apply for other parallel locality models like Y-PRAM, D-BSP and E-BSP, too.     

Method of developing all-optical parallel to serial converter exploiting Kerr non-linearity of medium

Different all-optical logic operations, memory blocks etc. are developed using the inherent parallelism of optics. In any digital communication and computation system, the role of serial to parallel and parallel to serial data conversion are very much essential for making a data convenient for transition.Here in this paper, the authors propose a new scheme for developing an all-optical parallel to serial data conversion system by using optical J-K Flip-Flops and some logic gates based on optical non-linear switches. The uses of such switches are widely established in the area of optical parallel computation.This process can be extend for developing some all-optical digital devices like shift register, optical pulse counters etc. The proposed scheme deals with the best use of high parallelism of the optics, so the super-fast processing speed can be achieved.     

Large-scale forcing with less communication in finite-difference simulations of stationary isotropic turbulence

This study proposes a new forcing scheme suitable for massively-parallel finite-difference simulations of stationary isotropic turbulence. The proposed forcing scheme, named reduced-communication forcing (RCF), is based on the same idea as the conventional large-scale forcing scheme, but requires much less data communication, leading to a high parallel efficiency. It has been confirmed that the RCF scheme works intrinsically in the same manner as the conventional large-scale forcing scheme. Comparisons have revealed that a fourth-order finite-difference model run in combination with the RCF scheme (FDM-RCF) is as good as a spectral model, while requiring less computational costs. For the range 80 < Re_λ < 540, where Re_λ is the Taylor microscale-based Reynolds number, large computations using the FDM-RCF scheme show that the Reynolds dependences of skewness and flatness factors have similar power-laws as found in previous studies.     

Mixture diffusion of sulfonated dyes into cellulose membrane. III. Application of parallel diffusion model to binary system of direct dyes with different affinity

Mixture diffusion of two dyes (C.I. Direct Blue 15 (DB15) and C.I. Direct Yellow 12 (DY12)) with different affinity onto the substrate into cellulose membrane from the binary solution was studied at 55°C. Uptake curves and concentration–distance profiles were measured experimentally in the ratios (DB15:DY12) 1:0.5, 1:1 and 1:2. It was examined whether the diffusion of the dyes could be analyzed based on the parallel diffusion theory of surface and pore diffusion. It was revealed that the diffusion of DB15 with higher affinity could be analyzed based on the model in the ratios 1:0.5 and 1:1, although the theoretical value deviated slightly from the data in the concentration–distance profile in the ratio 1:1. On the other hand, the diffusion of DY12 with smaller affinity could not be described by the model, because the diffusivity of the dye changed during the adsorption process against the assumption of the model.     

Minimizing the number of machines for minimum length schedules

In this paper, we present a new objective function for scheduling on parallel machines: minimizing the number of machines for schedules of minimum length. We study its complexity and we prove the NP-completeness of this problem, even if there is no precedences or for unitary execution times. We propose several polynomial algorithms for various particular cases.     

Probabilistically induced domain decomposition methods for elliptic boundary-value problems

Monte Carlo as well as quasi-Monte Carlo methods are used to generate only few interfacial values in two-dimensional domains where boundary-value elliptic problems are formulated. This allows for a domain decomposition of the domain. A continuous approximation of the solution is obtained interpolating on such interfaces, and then used as boundary data to split the original problem into fully decoupled subproblems. The numerical treatment can then be continued, implementing any deterministic algorithm on each subdomain. Both, Monte Carlo (or quasi-Monte Carlo) simulations and the domain decomposition strategy allow for exploiting parallel architectures. Scalability and natural fault tolerance are peculiarities of the present algorithm. Examples concern Helmholtz and Poisson equations, whose probabilistic treatment presents additional complications with respect to the case of homogeneous elliptic problems without any potential term and source.     

Four-operand parallel optical computing using shadow-casting technique

Optical shadow-casting (OSC) technique has shown excellent potential for optically implementing two-operand parallel logic gates and array logic operations. The 16 logic functions for two binary patterns (variables) are optically realizable in parallel by properly configuring an array of 2×2 light emitting diodes. In this paper, we propose an enhanced OSC technique for implementing four-operand parallel logic gates. The proposed system is capable of performing 2¹⁶ logic functions by simply programming the switching mode of an array of 4×4 light emitting diodes in the input plane. This leads to an efficient and compact realization scheme when compared to the conventional two-operand OSC system.