Pavel Karas

A Benchmark for Comparison of Cell Tracking Algorithms

Motivation: Automatic tracking of cells in multidimensional time-lapse fluorescence microscopy is an important task in many biomedical applications. A novel framework for objective evaluation of cell tracking algorithms has been established under the auspices of the IEEE International Symposium on Biomedical Imaging 2013 Cell Tracking Challenge. In this article, we present the logistics, datasets, methods and results of the challenge and lay down the principles for future uses of this benchmark. Results: The main contributions of the challenge include the creation of a comprehensive video dataset repository and the definition of objective measures for comparison and ranking of the algorithms. With this benchmark, six algorithms covering a variety of segmentation and tracking paradigms have been compared and ranked based on their performance on both synthetic and real datasets. Given the diversity of the datasets, we do not declare a single winner of the challenge. Instead, we present and discuss the results for each individual dataset separately. Availability and implementation: The challenge Web site (http://www.codesolorzano.com/celltrackingchallenge) provides access to the training and competition datasets, along with the ground truth of the training videos. It also provides access to Windows and Linux executable files of the evaluation software and most of the algorithms that competed in the challenge. Contact: codesolorzano@unav.es Supplementary information: Supplementary data are available at Bioinformatics online.

Convolution of large 3D images on GPU and its decomposition

In this article, we propose a method for computing convolution of large 3D images. The convolution is performed in a frequency domain using a convolution theorem. The algorithm is accelerated on a graphic card by means of the CUDA parallel computing model. Convolution is decomposed in a frequency domain using the decimation in frequency algorithm. We pay attention to keeping our approach efficient in terms of both time and memory consumption and also in terms of memory transfers between CPU and GPU which have a significant inuence on overall computational time. We also study the implementation on multiple GPUs and compare the results between the multi-GPU and multi-CPU implementations.

Algorithms for Efficient Computation of Convolution

Convolution is an important mathematical tool in both fields of signal and image processing. It is em-ployed in filtering, denoising, edge detection, correlation, compression, deconvolution, simulation, and in many other applications. Although the concept of convolution is not new, the efficient computation of convolution is still an open topic. As the amount of processed data is constantly increasing, there is considerable request for fast manipulation with huge data. Moreover, there is demand for fast algorithms which can exploit computational power of modern parallel architectures. The aim of this chapter is to review the algorithms and approaches for computation of convolution with regards to various properties such as signal and kernel size or kernel separability (when pro-cessing n-dimensional signals). Target architectures include superscalar and parallel processing units (namely CPU, DSP, and GPU), programmable architectures (e.g. FPGA), and distributed systems (such as grids). The structure of the chapter is designed to cover various applications with respect to the signal size, from small to large scales.

Simple non-invasive analysis of embryonic stem cell-derived cardiomyocytes beating in vitro.

The analysis of digital video output enables the non-invasive screening of various active biological processes. For the monitoring and computing of the beating parameters of cardiomyocytes in vitro, CB Analyser (cardiomyocyte beating analyser) software was developed. This software is based on image analysis of the video recording of beating cardiomyocytes. CB Analyser was tested using cardiomyocytes derived from mouse embryonic stem cells at different stages of cardiomyogenesis. We observed that during differentiation (from day 18), the beat peak width decreased, which corresponded to the increased speed of an individual pulse. However, the beating frequency did not change. Further, the effects of epinephrine modulating mature cardiomyocyte functions were tested to validate the CB Analyser analysis. In conclusion, data show that CB Analyser is a useful tool for evaluating the functions of both developing and mature cardiomyocytes under various conditions in vitro.

GPU implementation of linear morphological openings with arbitrary angle

Linear morphological openings and closings are important non-linear operators from mathematical morphology. In practical applications, many different orientations of digital line segments must typically be considered. In this paper, we (1) review efficient sequential as well as parallel algorithms for the computation of linear openings and closings; (2) compare the performance of CPU implementations of four state-of-the-art algorithms; (3) describe GPU implementations of two recent efficient algorithms allowing arbitrary orientation of the line segments; (4) propose, as the main contribution, an efficient and optimized GPU implementation of linear openings; and (5) compare the performance of all implementations on real images from various applications. From our experimental results, it turned out that the proposed GPU implementation is suitable for applications with large, industrial images, running under severe timing constraints.

GPU optimization of convolution for large 3-d real images

In this paper, we propose a method for computing convolution of large 3-D images with respect to real signals. The convolution is performed in a frequency domain using a convolution theorem. Due to properties of real signals, the algorithm can be optimized so that both time and the memory consumption are halved when compared to complex signals of the same size. Convolution is decomposed in a frequency domain using the decimation in frequency (DIF) algorithm. The algorithm is accelerated on a graphics hardware by means of the CUDA parallel computing model, achieving up to 10× speedup with a single GPU over an optimized implementation on a quad-core CPU.

A simple microviscometric approach based on Brownian motion tracking

Viscosity-an integral property of a liquid-is traditionally determined by mechanical instruments. The most pronounced disadvantage of such an approach is the requirement of a large sample volume, which poses a serious obstacle, particularly in biology and biophysics when working with limited samples. Scaling down the required volume by means of microviscometry based on tracking the Brownian motion of particles can provide a reasonable alternative. In this paper, we report a simple microviscometric approach which can be conducted with common laboratory equipment. The core of this approach consists in a freely available standalone script to process particle trajectory data based on a Newtonian model. In our study, this setup allowed the sample to be scaled down to 10 μl. The utility of the approach was demonstrated using model solutions of glycerine, hyaluronate, and mouse blood plasma. Therefore, this microviscometric approach based on a newly developed freely available script can be suggested for determination of the viscosity of small biological samples (e.g., body fluids).

Deconvolution of Huge 3-D Images: Parallelization Strategies on a Multi-GPU System

In this paper, we discuss strategies to parallelize selected deconvolution methods on a multi-GPU system. We provide a comparison of several approaches to split the deconvolution into subtasks while keeping the amount of costly data transfers as low as possible, and propose own implementation of three deconvolution methods which achieves up to 65× speedup over the CPU one. In the experimental part, we analyse how the individual stages of the computation contribute to the overall computation time as well as how the multi-GPU implementation scales in various setups. Finally, we identify bottlenecks of the system.