Limits...
Fast multi-core based multimodal registration of 2D cross-sections and 3D datasets.

Scharfe M, Pielot R, Schreiber F - BMC Bioinformatics (2010)

Bottom Line: We evaluate the CBE-driven PlayStation 3 as a high performance, cost-effective computing platform by adapting a multimodal alignment procedure to several characteristic hardware properties.The optimisations are based on partitioning, vectorisation, branch reducing and loop unrolling techniques with special attention to 32-bit multiplies and limited local storage on the computing units.We discuss several CBE-based optimisation methods and compare our results to standard solutions.

View Article: PubMed Central - HTML - PubMed

Affiliation: Leibniz-Institute of Plant Genetics and Crop Plant Research (IPK), Gatersleben, Germany.

ABSTRACT

Background: Solving bioinformatics tasks often requires extensive computational power. Recent trends in processor architecture combine multiple cores into a single chip to improve overall performance. The Cell Broadband Engine (CBE), a heterogeneous multi-core processor, provides power-efficient and cost-effective high-performance computing. One application area is image analysis and visualisation, in particular registration of 2D cross-sections into 3D image datasets. Such techniques can be used to put different image modalities into spatial correspondence, for example, 2D images of histological cuts into morphological 3D frameworks.

Results: We evaluate the CBE-driven PlayStation 3 as a high performance, cost-effective computing platform by adapting a multimodal alignment procedure to several characteristic hardware properties. The optimisations are based on partitioning, vectorisation, branch reducing and loop unrolling techniques with special attention to 32-bit multiplies and limited local storage on the computing units. We show how a typical image analysis and visualisation problem, the multimodal registration of 2D cross-sections and 3D datasets, benefits from the multi-core based implementation of the alignment algorithm. We discuss several CBE-based optimisation methods and compare our results to standard solutions. More information and the source code are available from http://cbe.ipk-gatersleben.de.

Conclusions: The results demonstrate that the CBE processor in a PlayStation 3 accelerates computational intensive multimodal registration, which is of great importance in biological/medical image processing. The PlayStation 3 as a low cost CBE-based platform offers an efficient option to conventional hardware to solve computational problems in image processing and bioinformatics.

Show MeSH

Related in: MedlinePlus

A volume rendering of the 3D NMR dataset of a brain (see Figure 9) together with a registered 2D PET scan (see Figure 10) after the multimodal alignment procedure.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2821319&req=5

Figure 1: A volume rendering of the 3D NMR dataset of a brain (see Figure 9) together with a registered 2D PET scan (see Figure 10) after the multimodal alignment procedure.

Mentions: Comprehensive understanding of biological structures requires sophisticated techniques in many areas such as the combination of 2D and 3D images or models of biological objects. Examples are the integration of histological cross-sections providing structural information and development-specific distribution patterns of mRNA, metabolite concentrations or enzyme activities into a 3D morphological framework [1], the combination of 2D computer tomography (CT) slices with a 3D atlas [2], or the integration of 2D positron emission tomography (PET) slices, providing information about metabolic activity, into a 3D NMR dataset, see Figure 1. For reconstruction and 3D visualisation these 2D cross-sections have to be registered at correct spatial positions in a 3D morphological framework. Manual registration of cross-sections is tedious, subjective and very time-consuming. The accurate registration of images, obtained by diverse imaging techniques, requires automatic multimodal alignment techniques, which is an important research field in biological and medical image processing [3-5]. Such approaches determine the optimal spatial position on the base of suitable similarity functions, such as cross correlation [6] or mutual information [3,7,8]. Registration by automatic procedures still requires extensive computational resources, so that fast algorithms and the implementation on parallel hardware would greatly enhance the feasibility of these investigations. The multi-core Cell Broadband Engine (CBE) allows fast parallel computation on eight cores per chip, presenting potential as a target for implementation of image registration algorithms [9,10]. In this paper we implemented and evaluated a multimodal alignment approach based on mutual information on the CBE. The availability of the inexpensive CBE-driven PlayStation 3 provides the opportunity to simultaneously align a high number of image stacks on a low-cost platform and therefore improves the automatic analysis and visualisation of biological information obtained through diverse imaging methods. We discuss several CBE-based optimisation methods and compare our results to standard solutions.


Fast multi-core based multimodal registration of 2D cross-sections and 3D datasets.

Scharfe M, Pielot R, Schreiber F - BMC Bioinformatics (2010)

A volume rendering of the 3D NMR dataset of a brain (see Figure 9) together with a registered 2D PET scan (see Figure 10) after the multimodal alignment procedure.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2821319&req=5

Figure 1: A volume rendering of the 3D NMR dataset of a brain (see Figure 9) together with a registered 2D PET scan (see Figure 10) after the multimodal alignment procedure.
Mentions: Comprehensive understanding of biological structures requires sophisticated techniques in many areas such as the combination of 2D and 3D images or models of biological objects. Examples are the integration of histological cross-sections providing structural information and development-specific distribution patterns of mRNA, metabolite concentrations or enzyme activities into a 3D morphological framework [1], the combination of 2D computer tomography (CT) slices with a 3D atlas [2], or the integration of 2D positron emission tomography (PET) slices, providing information about metabolic activity, into a 3D NMR dataset, see Figure 1. For reconstruction and 3D visualisation these 2D cross-sections have to be registered at correct spatial positions in a 3D morphological framework. Manual registration of cross-sections is tedious, subjective and very time-consuming. The accurate registration of images, obtained by diverse imaging techniques, requires automatic multimodal alignment techniques, which is an important research field in biological and medical image processing [3-5]. Such approaches determine the optimal spatial position on the base of suitable similarity functions, such as cross correlation [6] or mutual information [3,7,8]. Registration by automatic procedures still requires extensive computational resources, so that fast algorithms and the implementation on parallel hardware would greatly enhance the feasibility of these investigations. The multi-core Cell Broadband Engine (CBE) allows fast parallel computation on eight cores per chip, presenting potential as a target for implementation of image registration algorithms [9,10]. In this paper we implemented and evaluated a multimodal alignment approach based on mutual information on the CBE. The availability of the inexpensive CBE-driven PlayStation 3 provides the opportunity to simultaneously align a high number of image stacks on a low-cost platform and therefore improves the automatic analysis and visualisation of biological information obtained through diverse imaging methods. We discuss several CBE-based optimisation methods and compare our results to standard solutions.

Bottom Line: We evaluate the CBE-driven PlayStation 3 as a high performance, cost-effective computing platform by adapting a multimodal alignment procedure to several characteristic hardware properties.The optimisations are based on partitioning, vectorisation, branch reducing and loop unrolling techniques with special attention to 32-bit multiplies and limited local storage on the computing units.We discuss several CBE-based optimisation methods and compare our results to standard solutions.

View Article: PubMed Central - HTML - PubMed

Affiliation: Leibniz-Institute of Plant Genetics and Crop Plant Research (IPK), Gatersleben, Germany.

ABSTRACT

Background: Solving bioinformatics tasks often requires extensive computational power. Recent trends in processor architecture combine multiple cores into a single chip to improve overall performance. The Cell Broadband Engine (CBE), a heterogeneous multi-core processor, provides power-efficient and cost-effective high-performance computing. One application area is image analysis and visualisation, in particular registration of 2D cross-sections into 3D image datasets. Such techniques can be used to put different image modalities into spatial correspondence, for example, 2D images of histological cuts into morphological 3D frameworks.

Results: We evaluate the CBE-driven PlayStation 3 as a high performance, cost-effective computing platform by adapting a multimodal alignment procedure to several characteristic hardware properties. The optimisations are based on partitioning, vectorisation, branch reducing and loop unrolling techniques with special attention to 32-bit multiplies and limited local storage on the computing units. We show how a typical image analysis and visualisation problem, the multimodal registration of 2D cross-sections and 3D datasets, benefits from the multi-core based implementation of the alignment algorithm. We discuss several CBE-based optimisation methods and compare our results to standard solutions. More information and the source code are available from http://cbe.ipk-gatersleben.de.

Conclusions: The results demonstrate that the CBE processor in a PlayStation 3 accelerates computational intensive multimodal registration, which is of great importance in biological/medical image processing. The PlayStation 3 as a low cost CBE-based platform offers an efficient option to conventional hardware to solve computational problems in image processing and bioinformatics.

Show MeSH
Related in: MedlinePlus