Oleh Dzyubachyk

Methods for cell and particle tracking.

Achieving complete understanding of any living thing inevitably requires thorough analysis of both its anatomic and dynamic properties. Live-cell imaging experiments carried out to this end often produce massive amounts of time-lapse image data containing far more information than can be digested by a human observer. Computerized image analysis offers the potential to take full advantage of available data in an efficient and reproducible manner. A recurring task in many experiments is the tracking of large numbers of cells or particles and the analysis of their (morpho)dynamic behavior. In the past decade, many methods have been developed for this purpose, and software tools based on these are increasingly becoming available. Here, we survey the latest developments in this area and discuss the various computational approaches, software tools, and quantitative measures for tracking and motion analysis of cells and particles in time-lapse microscopy images.

Advanced Level-Set-Based Cell Tracking in Time-Lapse Fluorescence Microscopy

Cell segmentation and tracking in time-lapse fluorescence microscopy images is a task of fundamental importance in many biological studies on cell migration and proliferation. In recent years, level sets have been shown to provide a very appropriate framework for this purpose, as they are well suited to capture topological changes occurring during mitosis, and they easily extend to higher dimensional image data. This model evolution approach has also been extended to deal with many cells concurrently. Notwithstanding its high potential, the multiple-level-set method suffers from a number of shortcomings, which limit its applicability to a larger variety of cell biological imaging studies. In this paper, we propose several modifications and extensions to the coupled-active-surfaces algorithm, which considerably improve its robustness and applicability. Our algorithm was validated by comparing it to the original algorithm and two other cell segmentation algorithms. For the evaluation, four real fluorescence microscopy image datasets were used, involving different cell types and labelings that are representative of a large range of biological experiments. Improved tracking performance in terms of precision (up to 11%), recall (up to 8%), ability to correctly capture all cell division events, and computation time (up to nine times reduction) is achieved.

Tracking in cell and developmental biology.

The past decade has seen an unprecedented data explosion in biology. It has become evident that in order to take full advantage of the potential wealth of information hidden in the data produced by even a single experiment, visual inspection and manual analysis are no longer adequate. To ensure efficiency, consistency, and completeness in data processing and analysis, computational tools are essential. Of particular importance to many modern live-cell imaging experiments is the ability to automatically track and analyze the motion of objects in time-lapse microscopy images. This article surveys the recent literature in this area. Covering all scales of microscopic observation, from cells, down to molecules, and up to entire organisms, it discusses the latest trends and successes in the development and application of computerized tracking methods in cell and developmental biology.

Advanced level-set based multiple-cell segmentation and tracking in time-lapse fluorescence microscopy images

Segmentation and tracking of cells in fluorescence microscopy image sequences is an important task in many biological studies into cell migration as well as intracellular dynamics. The growing size and complexity of biological image data sets precludes manual analysis, and calls for increasingly advanced automatic algorithms that are generic enough to be easily tunable to different applications, yet robust enough to deal with different cell types and strongly varying imaging conditions. Active-contour based algorithms have proven to be very suitable for this purpose but still suffer from several shortcomings that limit their segmentation accuracy and tracking robustness. In addition, these algorithms are generally rather computationally expensive. In this paper, we present an advanced level-set based multiple-cell segmentation and tracking algorithm, which implements seven modifications compared to earlier algorithms that considerably improve its performance. Preliminary experiments on three different time-lapse fluorescence microscopy images demonstrate the potential of the new algorithm.

Chapter 15 – Time‐Lapse Imaging

Publisher Summary nTime-lapse imaging experiments involve the acquisition of not only spatial information, but also temporal information, and often-spectral information as well, resulting in up to five-dimensional image data sets. The ultimate goal of time-lapse imaging experiments is to gain an insight into cellular and intracellular dynamic processes. Inevitably this requires quantitative analysis of motion patterns. Approaches to this problem fall into three categories. The first consists of real-time, single-target tracking techniques. These usually involve a microscope setup containing an image-based feedback loop controlling the positioning and focusing of the system to keep the object of interest in the center of the field of view. Only a small portion of the specimen is illuminated this way, which reduces photo damage and allows imaging to be done faster or over a longer time. The second category consists of ensemble tracking approaches, such as fluorescence recovery after photobleaching and fluorescence loss in photobleaching. While useful for assessing specific dynamic parameters (e.g., diffusion coefficients and association and dissociation rates of labeled proteins), they are limited to yielding averages over larger populations. The third category consists of approaches that aim to track all individual objects of interest present in the data. These are usually performed offline.

Comprehensive single cell-resolution analysis of the role of chromatin regulators in early C. elegans embryogenesis

Chromatin regulators are widely expressed proteins with diverse roles in gene expression, nuclear organization, cell cycle regulation, pluripotency, physiology and development, and are frequently mutated in human diseases such as cancer. Their inhibition often results in pleiotropic effects that are difficult to study using conventional approaches. We have developed a semi-automated nuclear tracking algorithm to quantify the divisions, movements and positions of all nuclei during the early development of Caenorhabditis elegans and have used it to systematically study the effects of inhibiting chromatin regulators. The resulting high dimensional datasets revealed that inhibition of multiple regulators, including F55A3.3 (encoding FACT subunit SUPT16H), lin-53 (RBBP4/7), rba-1 (RBBP4/7), set-16 (MLL2/3), hda-1 (HDAC1/2), swsn-7 (ARID2), and let-526 (ARID1A/1B) affected cell cycle progression and caused chromosome segregation defects. In contrast, inhibition of cir-1 (CIR1) accelerated cell division timing in specific cells of the AB lineage. The inhibition of RNA polymerase II also accelerated these division timings, suggesting that normal gene expression is required to delay cell cycle progression in multiple lineages in the early embryo. Quantitative analyses of the dataset suggested the existence of at least two functionally distinct SWI/SNF chromatin remodeling complex activities in the early embryo, and identified a redundant requirement for the egl-27 and lin-40 MTA orthologs in the development of endoderm and mesoderm lineages. Moreover, our dataset also revealed a characteristic rearrangement of chromatin to the nuclear periphery upon the inhibition of multiple general regulators of gene expression. Our systematic, comprehensive and quantitative datasets illustrate the power of single cell-resolution quantitative tracking and high dimensional phenotyping to investigate gene function. Furthermore, the results provide an overview of the functions of essential chromatin regulators during the early development of an animal.

Model-based approach for tracking embryogenesis in Caenorhabditis elegans fluorescence microscopy data

The nematode Caenorhabditis elegans (C. elegans) is a widely used model organism in biological investigations. Due to its well-known and invariant cell lineage tree, it can be used to study the effects of mutations and various disease processes. Effective and efficient analysis of the wealth of time-lapse fluorescence microscopy image data acquired in such studies requires automation of the cell segmentation and tracking tasks involved. This is hampered by many factors, including autofluorescence effects, low and uneven contrast throughout the images, high noise levels, large numbers of possibly simultaneous cell divisions, and touching or clustering cells. In this paper, we present a new algorithm for segmentation and tracking of cells in C. elegans embryogenesis image data. It is based on the model evolution framework for image segmentation and uses a novel multi-object tracking scheme based on energy minimization via graph cuts. Preliminary experiments on publicly available test data demonstrate the potential of the algorithm compared to existing approaches.

A VARIATIONAL MODEL FOR LEVEL-SET BASED CELL TRACKING IN TIME-LAPSE FLUORESCENCE MICROSCOPY IMAGES

Quantifying the motion and deformation of large numbers of cells through image sequences obtained with fluorescence microscopy is a recurrent task in many biological studies. Automated segmentation and tracking methods are increasingly needed to be able to analyze the large amounts of image data acquired for such studies. In addition, automated techniques have the possibility to improve sensitivity, objectivity, and reproducibility compared to human observers. Recent efforts in this area have revealed the potential of model evolution methods, notably active contours and level sets, for this purpose. One of the disadvantages of such methods is their sensitivity to parameter settings. In this paper we propose a variational model for level-set based cell tracking which involves less parameters with more intuitive meaning compared to previous approaches. The improved performance is demonstrated with experimental results on real time-lapse fluorescence microscopy image data

Automated lineage tree reconstruction from Caenorhabditis elegans image data using particle filtering based cell tracking

Caenorhabditis elegans is an important model organism for the study of molecular mechanisms of development and disease processes, due to its well-known genome and invariant cell lineage tree. Such studies generally produce vast amounts of image data, and require very robust and efficient algorithms to extract and characterize lineage phenotypes and to determine gene expression patterns. Previously published methods for this purpose show only mediocre performance and often require extensive manual post-processing. The challenge remains to develop more powerful and fully automated methods. In this paper we propose a new algorithm for C. elegans cell tracking and lineage reconstruction, based on a Bayesian estimation framework, implemented by means of particle filtering. The tracking is enhanced with a detection stage, based on the h-dome transform. Preliminary experiments on several image sequences demonstrate that the new tracking algorithm is able to reconstruct the lineage tree, at least until the 350-cell stage, without manual intervention, at low computational cost and with low error rates.

Energy minimization methods for cell motion correction and intracellular analysis in live-cell fluorescence microscopy

The ultimate aim of many live-cell fluorescence microscopy imaging experiments is the quantitative analysis of the spatial structure and temporal behavior of intracellular objects. This requires finding the precise geometrical correspondence between the time frames for each individual cell and performing intracellular segmentation. In a previous paper we have developed a powerful multi-level-set based algorithm for automated cell segmentation and tracking of many cells in time-lapse images. In this paper, we propose approaches to exploit the output of this algorithm for the subsequent tasks of cell motion correction and intracellular segmentation. Both tasks are formulated as energy minimization problems and are solved efficiently and effectively by distance-transform and graph-cut based algorithms. The potential of the proposed approaches for intracellular analysis is demonstrated by successful experiments on biological image data showing PCNA-foci and nucleoli in HeLa cells.