Application of an M-line-based backprojected filtration algorithm to triple-cone-beam helical CT

Modern information technologies are increasingly focused on the tasks related to the knowledge processing: intelligent Internet search systems and enterprise resource planning, DataMining and smart home systems, recommender systems for various purposes, etc. Such problems are specified by the availability of large arrays of complexly structured symbol information, a high degree of implicit parallelism, weak formalizability, and a low need for complex arithmetic operations. As a result, the solution of such problems using the off-the-shelf microprocessors is extremely inefficient. One of approaches to solving this problem is to develop a specialized hardware device designed for high-performance symbolic data processing. The proposed logical inference engine is focused on the accelerated method of parallel logical inference that has a high degree of parallelism, and dataflow architecture. The article describes the development of the executive processor of the logical inference engine. The emphasize is placed on the unification unit design. A new method of parallel unification of the predicate pairs is presented. The unification unit logic diagrams developed using the Quartus CAD system have been implemented on FPGA by Altera. The comparative analysis of the sequential and parallel unit implementation on the basis of the experiments performed is provided.

Approximate string matching has been widely used in many applications, including deoxyribonucleic acid sequence searching, spell checking, text mining, and spam filters. The method is designed to find all locations of strings that approximately match a pattern in accordance with the number of insertion, deletion, and substitution operations. Among the proposed algorithms, the bit-parallel algorithms are considered to be the best and highly efficient algorithms. However, the traditional bit-parallel algorithms lacks the ability of identifying the start and end positions of a matched pattern. Furthermore, acceleration of the bit-parallel algorithms has become a crucial issue for processing big data nowadays. In this paper, we propose two kinds of parallel location-aware algorithms called data-segmented parallelism and high-degree parallelism as means to accelerate approximate string matching using graphic processing units. Experimental results show that the high-degree parallelism on GPUs achieves significant improvement in system and kernel throughputs compared to CPU counterparts. Compared to state-of-the-art approaches, the proposed high-degree parallelism achieves 11 to 105 times improvement.

Humanoid robots find application in human-robot interaction tasks. However, despite their capabilities, their sequential computing system limits the execution of computationally expensive algorithms such as convolutional neural networks, which have demonstrated good performance in recognition tasks. As an alternative to sequential computing units, Field-Programmable Gate Arrays and Graphics Processing Units have a high degree of parallelism and low power consumption. This study aims to improve the visual perception of a humanoid robot called NAO using these embedded systems running a convolutional neural network. The methodology adopted here is based on image acquisition and transmission using simulation software: Webots and Choreographe. In each embedded system, an object recognition stage is performed using commercial convolutional neural network acceleration frameworks. Xilinx® Ultra96™, Intel® Cyclone® V-SoC and NVIDIA® Jetson™ TX2 cards were used, and Tinier-YOLO, AlexNet, Inception-V1 and Inception V3 transfer-learning networks were executed. Real-time metrics were obtained when Inception V1, Inception V3 transfer-learning and AlexNet were run on the Ultra96 and Jetson TX2 cards, with frame rates between 28 and 30 frames per second. The results demonstrated that the use of these embedded systems and convolutional neural networks can provide humanoid robots such as NAO with greater visual recognition in tasks that require high accuracy and autonomy.

Projected and subspace clustering aim to find groups of similar objects within a subspace of the full-dimensional space. Where subspace clustering tries to identify clusters in all possible subspaces, projected clustering assigns each point to a single cluster within one projected subspace, resulting in a much smaller result set. PROCLUS is an adaptation of the k-medoids clustering algorithm, CLARANS, to projected clustering. Even though PROCLUS is the first projected clustering algorithm, it is still competitive in comparative empirical studies. PROCLUS is, however, still too slow for large-scale data or real-time interaction when used in information retrieval processes. Therefore, we propose novel algorithmic strategies to reduce computations and exploit the massive parallelism offered by modern graphical processing units (GPUs). To take advantage of their high degree of parallelism, standard sequential algorithms need to be significantly restructured. We therefore also propose a novel GPU-parallelized algorithm, GPU-FAST-PROCLUS, that takes advantage of the computational power of modern GPUs. We provide experimental studies that demonstrate the benefit of our proposed strategies and GPU-parallelizations. In this experimental evaluation, we obtain 3 orders of magnitude speedup compared to PROCLUS.

"High-precision, reconstructed 3D model" of skull scanned by conebeam CT: Reproducibility verified using CAD/CAM data.

The growth of data to be processed in the Oil & Gas industry matches the requirements imposed by evolving algorithms based on stencil computations, such as Full Waveform Inversion and Reverse Time Migration. Graphical processing units (GPUs) are an attractive architectural target for stencil computations because of its high degree of data parallelism. However, the rapid architectural and technological progression makes it difficult for even the most proficient programmers to remain up-to-date with the technological advances at a micro-architectural level. In this work, we present an extension for an open source compiler designed to produce highly optimized finite difference kernels for use in inversion methods named Devito. We embed it with the Oxford Parallel Domain Specific Language (OP-DSL) in order to enable automatic code generation for GPU architectures from a high-level representation. We aim to enable users coding in a symbolic representation level to effortlessly get their implementations leveraged by the processing capacities of GPU architectures. The implemented backend is evaluated on a NVIDIA GTX Titan Z, and on a NVIDIA Tesla V100 in terms of operational intensity through the roof-line model for varying space-order discretization levels of 3D acoustic isotropic wave propagation stencil kernels with and without symbolic optimizations. It achieves approximately 63% of V100's peak performance and 24% of Titan Z's peak performance for stencil kernels over grids with 256 points. Our study reveals that improving memory usage should be the most efficient strategy for leveraging the performance of the implemented solution on the evaluated architectures.

In some applications, evolutionary algorithms may require high computational resources and high processing power, sometimes not producing a satisfactory solution after running for a considerable amount of time. One possible improvement is a parallel approach to reduce the response time. This work proposes to study a parallel multi-objective algorithm, the multi-objective version of Differential Evolution (DE). The generation of trial individuals can be done in parallel, greatly reducing the overall processing time of the algorithm. A novel approach to parallelize this algorithm is the implementation on the Graphic Processing Units (GPU). These units present high degree of parallelism and they were initially developed for image rendering. However, NVIDIA has released a framework, named CUDA, which allows developers to use GPU for general-purpose computing (GPGPU). This work studies the implementation of Multi-Objective DE (MODE) on the GPU with C-CUDA, evaluating the gain in processing time against the sequential version. Benchmark functions are used to validate the implementation and to confirm the efficiency of MODE on the GPU. The results show that the approach achieves an expressive speed up and a highly efficient processing power.

Development of a Compact PET for integrated PET/CT/RT to Streamline and Enhance Functional/Anatomic Image-Guided Preclinical Radiation Oncology Researches

This Human Skin Colour detection is analyzed to be a most suitable and strong cue for face detection and face recognition which finds various application in domain like surveillance, face biometrics, gesture recognition, interactive game application and many other. These applications require fast image processing in real time and demands enormously high performance with respect to time and processing speed. Earlier work done on sequential architecture does not provide required capability and can be achieved by parallel programming. As Image processing applications shows high degree of parallelism, they prove excellent source for multi-core platform. We can use GPU multithreaded parallel computing techniques to improve the speed of image processing. The paper proposes a data parallelism programming model for Human Skin Colour Detection algorithm. The objective is to increase the computational speed of the algorithm through data parallelism using CUDA framework. The framework is supported by OpenCV libraries and implemented by GPU (Graphics Processing Unit). The evaluation is done on basis of comparative analysis of serial and parallel programming computing. The pixel based Explicitly Defined Region skin classifier is used and RGB colour space is chosen due to its wider use in storing and processing of digital image data. To gain substantial acceleration in image computations, skin classifier code is off-loaded to GPU having compute capacity of 3.5 and rest of code is executed on CPU. We have tested module on static images, as well as on live camera captured images. SFA standard dataset is utilized to evaluate performance. Speedup of 23.45% was achieved using parallel programming.

A novel parallel algorithm is presented for generating random scale-free networks using the preferential attachment model. The algorithm, named cuPPA, is custom-designed for “single instruction multiple data (SIMD)” style of parallel processing supported by modern processors such as graphical processing units (GPUs). To the best of our knowledge, our algorithm is the first to exploit GPUs, and also the fastest implementation available today, to generate scale-free networks using the preferential attachment model. A detailed performance study is presented to understand the scalability and runtime characteristics of the cuPPA algorithm. Also another version of the algorithm called cuPPA-Hash tailored for multiple GPUs is presented. On a single GPU, the original cuPPA algorithm delivers the best performance, but is challenging to port to multi-GPU implementation. For multi-GPU implementation, cuPPA-Hash has been used as the parallel algorithm to achieve a perfect linear speedup up to 4 GPUs. In one of the best cases, when executed on an NVidia GeForce 1080 GPU, the original cuPPA generates a scale-free network of two billion edges in less than 3 s. On multi-GPU platforms, cuPPA-Hash generates a scale-free network of 16 billion edges in less than 7 s using a machine consisting of 4 NVidia Tesla P100 GPUs.

Redundant data weighting in helical CT affects reconstruction accuracy, susceptibility to motion artifacts and noise. Cone beam weighting (CBW) is used in helical CT with wide cone angle to reduce cone beam artifact and also improve detector utilization and reduce image noise. Earlier proposed approaches usually consider a case when reconstruction view range is less than one rotation (1-PI). However, most practical settings in clinical helical CT use between one and two rotations view range for each reconstructed slice. In this case CBW weighting can be extended to N-PI case. With N ≥ 2 streak artifacts are caused by singularity of the CBW function at Z-positions near the image plane. Note that in this case image plane projects to a small fraction of detector row; and the non-linear form of CBW weight cannot be accurately determined for all image pixels, which causes inaccuracies in CBW weighting that manifests itself in form of streak artifacts. We show that artifacts can be successfully removed by applying upsampling in the segment direction. By making the upsampling adaptive to the view Z-position, numerical performance of the proposed approach can be optimized. Evaluation shows that artifact is completely removed by the proposed approach.

This paper presents a parallel system capable of accelerating biological sequence alignment on the graphics processing unit (GPU) grid. The GPU grid in this paper is a desktop grid system that utilizes idle GPUs and CPUs in the office and home. Our parallel implementation employs a master-worker paradigm to accelerate an OpenGL-based algorithm that runs on a single GPU. We integrate this implementation into a screensaver-based grid system that detects idle resources on which the alignment code can run. We also show some experimental results comparing our implementation with three different implementations running on a single GPU, a single CPU, or multiple CPUs. As a result, we find that a single non-dedicated GPU can provide us almost the same throughput as two dedicated CPUs in our laboratory environment, where GPU-equipped machines are ordinarily used to develop GPU applications. In a dedicated environment, the GPU-accelerated code achieves five times higher throughput than the CPU-based code. Furthermore, a linear speedup of 30.7X is observed on a 32-node cluster of dedicated GPUs. We also implement a compute unified device architecture (CUDA) based algorithm to demonstrate further acceleration.

As so long, three-dimensional cone-beam computed tomography(CBCT) image reconstruction is a hot issue in medical imaging field. Often the computation operation of CBCT reconstruction is huge and the reconstruction time is long. Now with the development of computer technology, especially the rapid development of Graphics Processing Unit (GPU) based general-purpose computing technology enables fast CBCT reconstruction possible. In this paper, a CBCT reconstruction algorithm-ordered subset expectation maximization (OSEM) based on GPU is developed. Experiment on the Shepp-logan phantom model showed that a good speedup is received.

Purpose: To develop an adaptive radiotherapy (ART) strategy based on in vivo dose verification using volumetric imaging, real‐time physiologically constrained deformable image registration (PC‐DIR) and high performance computation (HPC). Method and Materials: Comparison of delivered and planned dose distributions for patients undergoing radiation therapy, especially IMRT, can be used to estimate treatment outcome, and develop a successful ART strategy that accounts for patient positioning errors, organ deformation, physiological changes (i.e. tumor shrinkage, edema), and incorporating this back into the treatment planning. Using 4DCT imaging, we have developed a ART methodology based on 1) PC‐DIR for auto‐contouring; 2) automated internal target volume generation; 3) probability density function (PDF) for specific structures; and 4) graphics processor unit (GPU) based HPC for real‐time computation using NVIDIA 8800GTX. Results: Compared with the same algorithm without the PC constraint, PC‐DIR resulted in superior performance with up to 91% reduction in inverse consistency errors. The computer generated ITV yielded contours typically within 3mm of physician drawn contours, and also served as a check of physician drawn ITVs. The PDF was constructed for the ITV to determine the dose distribution based on spatial probability using time averaging over a respiratory cycle. GPU based registrations of 256×256×64 imaging volumes required 3.6 sec for 100 iterations, a 56X speedup compared with conventional CPU computing (i.e. dual core 2.4GHz CPU). Dose remapping and dose volume histogram computations were near real‐time (⩽3sec). Conclusion: We have developed key elements of a clinically implementable ART strategy, which include real‐time computations using GPU, and accurate and robust registrations based on PC‐DIR. These elements incorporate 4DCT imaging coupled with GPU based PC‐DIR and frequent conebeam CT (CBCT) verification. This methodology can exploit ongoing improvement in GPU hardware performance and CBCT image quality.

Parallel accelerated Custom Correlation Coefficient calculations for genomics applications