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

Comparison between the mean computation times of one single alignment on the PS3 Cell Processor and a MPI-parallelised solution on an Opteron performed on the example datasets. Speedup (sp) compares each platform solution to the single-core solution on the Opteron. The average speedup of the PS3 Cell is 3.98 compared to the single-core Opteron, 1.99 compared to dual-core Opteron and 0.99 to quad-core Opteron solution.
© Copyright Policy - open-access
Related In: Results  -  Collection

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

Figure 12: Comparison between the mean computation times of one single alignment on the PS3 Cell Processor and a MPI-parallelised solution on an Opteron performed on the example datasets. Speedup (sp) compares each platform solution to the single-core solution on the Opteron. The average speedup of the PS3 Cell is 3.98 compared to the single-core Opteron, 1.99 compared to dual-core Opteron and 0.99 to quad-core Opteron solution.

Mentions: In this study, we used two 3D NMR datasets of the male and female brain, freely available from the Open Access Series of Imaging Studies (OASIS) project [24]. The dimensions of the 3D images were 256 × 175 × 176 voxel, an example of the data is shown in Figure 9. Three modified slices of each NMR datasets and three different 2D PET scans (see Figure 10 for an example), published by the National Institute of Aging [25], were used for registrations on the brain data. The 2D images were converted into gray-values and down-scaled to the respective resolution of the 3D dataset. Because of a given rough pre-alignment the search space could be constrained for the translation from -30 to +30 pixel and for the rotation-angle from -20° to +20°. Figure 1 shows an example of the multimodal registration of a 3D dataset (brain) and an associated 2D image (PET). The results of the analysis are detailed below and shown in Figures 11, 12 and 13.


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

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

Comparison between the mean computation times of one single alignment on the PS3 Cell Processor and a MPI-parallelised solution on an Opteron performed on the example datasets. Speedup (sp) compares each platform solution to the single-core solution on the Opteron. The average speedup of the PS3 Cell is 3.98 compared to the single-core Opteron, 1.99 compared to dual-core Opteron and 0.99 to quad-core Opteron solution.
© Copyright Policy - open-access
Related In: Results  -  Collection

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

Figure 12: Comparison between the mean computation times of one single alignment on the PS3 Cell Processor and a MPI-parallelised solution on an Opteron performed on the example datasets. Speedup (sp) compares each platform solution to the single-core solution on the Opteron. The average speedup of the PS3 Cell is 3.98 compared to the single-core Opteron, 1.99 compared to dual-core Opteron and 0.99 to quad-core Opteron solution.
Mentions: In this study, we used two 3D NMR datasets of the male and female brain, freely available from the Open Access Series of Imaging Studies (OASIS) project [24]. The dimensions of the 3D images were 256 × 175 × 176 voxel, an example of the data is shown in Figure 9. Three modified slices of each NMR datasets and three different 2D PET scans (see Figure 10 for an example), published by the National Institute of Aging [25], were used for registrations on the brain data. The 2D images were converted into gray-values and down-scaled to the respective resolution of the 3D dataset. Because of a given rough pre-alignment the search space could be constrained for the translation from -30 to +30 pixel and for the rotation-angle from -20° to +20°. Figure 1 shows an example of the multimodal registration of a 3D dataset (brain) and an associated 2D image (PET). The results of the analysis are detailed below and shown in Figures 11, 12 and 13.

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