Parallel medical image reconstruction: from graphics processing units (GPU) to Grids

Schellmann, Maraike; Gorlatch, Sergei; Meiländer, Dominik; Kösters, Thomas; Schäfers, Klaus P.; Wübbeling, Frank; Burger, Martin

doi:10.1007/s11227-010-0397-z

Cited by 20 publications

(12 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, cloud tracking (GrauerGray et al, 2008) [9], statistical static timing analysis (Gulati & Khatri, 2008) [10], biomedical image analysis (Hartley et al, 2008) [11], AES cryptography encoding and decoding (Manavski, 2007) [20], medical image construction (Schellmann, 2008) [28] and so on. These studies show promising results in computation time.…”

Section: Applications Of Gpu Computation To the Evolutionary Computationmentioning

confidence: 99%

Solving quadratic assignment problems by genetic algorithms with GPU computation

Tsutsui

Fujimoto

2009

Proceedings of the 11th Annual Conference Companion on Genetic and Evolutionary Computation Conference: Late Breaking Papers

View full text Add to dashboard Cite

This paper describes designing a parallel GA with GPU computation to solve the quadratic assignment problem (QAP) which is one of the hardest optimization problems in permutation domains. For the parallel method, a multiplepopulation, coarse-grained GA model was used. Each subpopulation is evolved by a multiprocessor in a GPU (NVIDIA GeForce GTX285). At predetermined intervals of generations all individuals in subpopulations are shuffled via the VRAM of the GPU. The instances on which this algorithm was tested were taken from the QAPLIB benchmark library. Results were promising, showing a speedup ration from 3 to 12 times, compared to the Intel i7 965 processor.

show abstract

Section: Applications Of Gpu Computation To the Evolutionary Computationmentioning

confidence: 99%

Solving quadratic assignment problems by genetic algorithms with GPU computation

Tsutsui

Fujimoto

2009

Proceedings of the 11th Annual Conference Companion on Genetic and Evolutionary Computation Conference: Late Breaking Papers

View full text Add to dashboard Cite

show abstract

“…One way to deal with this limitation is parallelizing (dividing in parts) image reconstruction. To this end and taking advantage of the fact that image reconstruction problems have a high degree of data parallelism and a large number of independent arithmetic calculations; Graphic Processing Unit (GPU) with the parallel programming model of Compute Unified Device Architecture (CUDA) [17]- [19] can be used to speed up image computational times.…”

Section: Introductionmentioning

confidence: 99%

Study of CT Images Processing with the Implementation of MLEM Algorithm using CUDA on NVIDIA’S GPU Framework

Valencia-Pérez

Hernández-López

Moreno-Barbosa

et al. 2020

J. Nucl. Phy. Mat. Sci. Rad. A.

View full text Add to dashboard Cite

In medicine, the acquisition process in Computed Tomography Images (CT) is obtained by a reconstruction algorithm. The classical method for image reconstruction is the Filtered Back Projection (FBP). This method is fast and simple but does not use any statistical information about the measurements. The appearance of artifacts and its low spatial resolution in reconstructed images must be considered. Furthermore, the FBP requires of optimal conditions of the projections and complete sets of data. In this paper a methodology to accelerate acquisition process for CT based on the Maximum Likelihood Estimation Method (MLEM) algorithm is presented. This statistical iterative reconstruction algorithm uses a GPU Programming Paradigms and was compared with sequential algorithms in which the reconstruction time was reduced by up to 3 orders of magnitude while preserving image quality. Furthermore, they showed a good performance when compared with reconstruction methods provided by commercial software. The system, which would consist exclusively of a commercial laptop and GPU could be used as a fast, portable, simple and cheap image reconstruction platform in the future.

show abstract

“…Proof of this trend is the high number of research works about the use of GPU computing applied to general purpose computation [33]. GPUs demonstrated to achieve high performance on solving complex problems on physics [5], medicine [40,41,44], and computer science [6,12,21]. Specifically, we focus on GPU implementations of data mining [24] and machine learning [45] algorithms.…”

Section: Introductionmentioning

confidence: 99%

Scalable CAIM discretization on multiple GPUs using concurrent kernels

2014

View full text Add to dashboard Cite

CAIM (Class-Attribute Interdependence Maximization) is one of the stateof-the-art algorithms for discretizing data for which classes are known. However, it may take a long time when run on high-dimensional large-scale data, with large number of attributes and/or instances. This paper presents a solution to this problem by introducing a GPU-based implementation of the CAIM algorithm that significantly speeds up the discretization process on big complex data sets. The GPU-based implementation is scalable to multiple GPU devices and enables the use of concurrent kernels execution capabilities of modern GPUs. The CAIM GPU-based model is evaluated and compared with the original CAIM using single and multi-threaded parallel configurations on 40 data sets with different characteristics. The results show great speedup, up to 139 times faster using 4 GPUs, which makes discretization of big data efficient and manageable. For example, discretization time of one big data set is reduced from 2 hours to less than 2 minutes.

show abstract

Parallel medical image reconstruction: from graphics processing units (GPU) to Grids

Cited by 20 publications

References 11 publications

Solving quadratic assignment problems by genetic algorithms with GPU computation

Solving quadratic assignment problems by genetic algorithms with GPU computation

Study of CT Images Processing with the Implementation of MLEM Algorithm using CUDA on NVIDIA’S GPU Framework

Scalable CAIM discretization on multiple GPUs using concurrent kernels

Contact Info

Product

Resources

About