Strategies for Implementing Hardware-Assisted High-Throughput Cellular Image Analysis

Abstract

R ecent advances in imaging technology for biomedicine, including high-speed microscopy, automated microscopy, and imaging flow cytometry are poised to have a large impact on clinical diagnostics, drug discovery, and biological research. Enhanced acquisition speed, resolution, and automation of sample handling are enabling researchers to probe biological phenomena at an increasing rate and achieve intuitive image-based results. However, the rich image sets produced by these tools are massive, possessing potentially millions of frames with tremendous depth and complexity. As a result, the tools introduce immense computational requirements, and, more importantly, the fact that image analysis operates at a much lower speed than image acquisition limits its ability to play a role in critical tasks in biomedicine such as real-time decision making. In this work, we present our strategy for high-throughput image analysis on a graphical processing unit platform. We scrutinized our original algorithm for detecting, tracking, and analyzing cell morphology in high-speed images and identified inefficiencies in image filtering and potential shortcut routines in the morphological analysis stage. Using our ''grid method'' for image enhancements resulted in an 8.54Â reduction in total run time, whereas origin centering allowed using a look up table for coordinate transformation, which reduced the total run time by 55.64Â. Optimization of parallelization and implementation of specialized image processing hardware will ultimately enable real-time analysis of high-throughput image streams and bring wider adoption of assays based on new imaging technologies. ( JALA 2011;16:422-30) INTRODUCTIONImaging is ubiquitous in industrial processing, medicine, environmental science, and cell biology. Given the diverse modes of imaging that exist, an image can contain a wealth of information about an object. Process quality control in semiconductor manufacturing and particle synthesis uses a number of spatial metrics from images from scanning electron microscopy, transmission electron microscopy, atomic force microscopy, and optical microscopy. 1,2 Imaging tools including positron emission tomography, X-ray, magnetic resonance imaging, and computed tomography are widely used in medicine for diagnostic and prognostic purposes. Ocean and waterway monitoring, a critical charge of environmental science, can be performed with high-speed camera-coupled flow cytometry whereby the diversity and density of microscopic organisms, key indicators of ecosystem health, can be identified. 3,4 In cell biology, for example, cell size, morphology, and location can be extracted from brightfield or phase-contrast images. And, the presence or location of biomolecules within cells can be obtained from fluorescence images of chemically labeled cells, which has recently been implemented with automated fluid handling and imagers for high-content analysis.5 As technology improves, imaging rates and resolution increase and the cost of acquiring image sets decreases ...

High-throughput single-microparticle imaging flow analyzer

“…High information content spatial metrics provided by imaging are used to characterize microscopic particles such as emulsions, microorganisms, and cells in particle synthesis, ecosystem monitoring, biofuel formulation, drug discovery, histopathology, and cytology-based diagnostics (3)(4)(5). An impressive arsenal of optical nanoscopy methods enables imaging of cellular structures beyond the diffraction limit of light (6)(7)(8).…”

High-Throughput Assessment of Cellular Mechanical Properties

“…Often this is done subsequent to the experiment and requires 10-15 minutes of high-end computation time. New hardware-based image-analysis implementations promise to eliminate this bottleneck (138,139).…”

A hardware accelerated approach for imaging flow cytometry

“…In particular, in imaging cytometry technology to analyze cells and micro particles [2,3]. Buschke et al present a cell analysis system for image cytometry using a DSP and FPGA [3].…”

“…Tse et al use a GPU-based system to accelerate cell analysis [2]. It parallelizes the morphological analysis that maps a pair of images at two different coordinates and bottomhat filters for image contrast enhancement.…”

“…This information can be used to characterize and sort the cell. Typically cells are analyzed based on some morphological paramters, such as their initial diameter, circularity, and deformability [2]. These characteristics can be easily derived if we know the exact radius at each point of the cell membrane.…”