Dirk R. Padfield

Spatio-temporal cell cycle phase analysis using level sets and fast marching methods

Enabled by novel molecular markers, fluorescence microscopy enables the monitoring of multiple cellular functions using live cell assays. Automated image analysis is necessary to monitor such model systems in a high-throughput and high-content environment. Here, we demonstrate the ability to simultaneously track cell cycle phase and cell motion at the single cell level. Using a recently introduced cell cycle marker, we present a set of image analysis tools for automated cell phase analysis of live cells over extended time periods. Our model-based approach enables the characterization of the four phases of the cell cycle G1, S, G2, and M, which enables the study of the effect of inhibitor compounds that are designed to block the replication of cancerous cells in any of the phases. We approach the tracking problem as a spatio-temporal volume segmentation task, where the 2D slices are stacked into a volume with time as the z dimension. The segmentation of the G2 and S phases is accomplished using level sets, and we designed a model-based shape/size constraint to control the evolution of the level set. Our main contribution is the design of a speed function coupled with a fast marching path planning approach for tracking cells across the G1 phase based on the appearance change of the nuclei. The viability of our approach is demonstrated by presenting quantitative results on both controls and cases in which cells are treated with a cell cycle inhibitor.

Coupled minimum-cost flow cell tracking for high-throughput quantitative analysis.

A growing number of screening applications require the automated monitoring of cell populations in a high-throughput, high-content environment. These applications depend on accurate cell tracking of individual cells that display various behaviors including mitosis, merging, rapid movement, and entering and leaving the field of view. Many approaches to cell tracking have been developed in the past, but most are quite complex, require extensive post-processing, and are parameter intensive. To overcome such issues, we present a general, consistent, and extensible tracking approach that explicitly models cell behaviors in a graph-theoretic framework. We introduce a way of extending the standard minimum-cost flow algorithm to account for mitosis and merging events through a coupling operation on particular edges. We then show how the resulting graph can be efficiently solved using algorithms such as linear programming to choose the edges of the graph that observe the constraints while leading to the lowest overall cost. This tracking algorithm relies on accurate denoising and segmentation steps for which we use a wavelet-based approach that is able to accurately segment cells even in images with very low contrast-to-noise. In addition, the framework is able to measure and correct for microscope defocusing and stage shift. We applied the algorithms on nearly 6000 images of 400,000 cells representing 32,000 tracks taken from five separate datasets, each composed of multiple wells. Our algorithm was able to segment and track cells and detect different cell behaviors with an accuracy of over 99%. This overall framework enables accurate quantitative analysis of cell events and provides a valuable tool for high-throughput biological studies.

Color and texture based segmentation of molecular pathology images usING HSOMS

Prostate cancer is the most common cancer among men, excluding skin cancer. It is diagnosed by histopathology interpretation of Hematoxylin and Eosin (H&E)-stained tissue sections. Gland and nuclei distributions vary with the disease grade, and the morphological features vary with the advance of cancer. A tissue microarray with known disease stages can be used to enable efficient pathology slide image analysis. We focus on an intuitive approach for segmenting such images, using the Hierarchical Self-Organizing Map (HSOM). Our approach introduces the use of unsupervised clustering using both color and texture features, and the use of unsupervised color merging outside of the HSOM framework. The HSOM was applied to segment 109 tissues composed of four tissue clusters: glands, epithelia, stroma, and nuclei. These segmentations were compared with the results of an EM Gaussian clustering algorithm. The proposed method confirms that the self-learning ability and adaptability of the HSOM, coupled with the information fusion mechanism of the hierarchical network, leads to superior segmentation results for tissue images.

Spatio-temporal cell segmentation and tracking for automated screening

A growing number of screening applications require the automated monitoring of cell populations including cell segmentation, tracking, and measurement. We present general methods for cell segmentation and tracking that exploit the spatio- temporal nature of the task to constrain segmentation. The images are de-noised and segmented by combining wavelet coefficients at various levels, thus enabling extraction of cells in images with low contrast-to-noise ratios. Each track of clustered cells resulting from association of nearby cells in the spatio-temporal volume is then split into individual cells by evolving sets of contours from other slices. The hypothesis whether to split or merge objects making up the cluster is tested using learned features trained from single track cells. Due to the difficult nature of generating ground truth, we also present a framework for edit-based validation whereby the user corrects the edits made by the automatic system rather than generating the truth from scratch. The results show the promise of the approach and demonstrate the ability of the algorithms to provide meaningful measurements of cell response to drug treatment in low-dose Hoechst-stained cells.

Coupled Minimum-Cost Flow Cell Tracking

A growing number of screening applications require the automated monitoring of cell populations in a high-throughput, high-content environment. These applications depend on accurate cell tracking of individual cells that display various behaviors including mitosis, occlusion, rapid movement, and entering and leaving the field of view. We present a tracking approach that explicitly models each of these behaviors and represents the association costs in a graph-theoretic minimum-cost flow framework. We show how to extend the minimum-cost flow algorithm to account for mitosis and merging events by coupling particular edges. We applied the algorithm to nearly 6,000 images of 400,000 cells representing 32,000 tracks taken from five separate datasets, each composed of multiple wells. Our algorithm is able to track cells and detect different cell behaviors with an accuracy of over 99%.

Spatio-temporal cell cycle analysis using 3D level set segmentation of unstained nuclei in line scan confocal fluorescence images

Automated analysis of live cells over extended time periods requires both novel assays and automated image analysis algorithms. Among other applications, this is necessary for studying the effect of inhibitor compounds which are designed to block the replication of cancerous cells in a high-throughput environment. Due to their toxicity, fluorescent dyes cannot be used to mark nuclei. Instead, the cell cycle itself may be marked with a fluorescent protein. This paper describes a set of image analysis methods designed to automatically segment nuclei in 2D time-lapse images. Since each nucleus is unstained, it needs to be segmented from the surrounding stained cytoplasm, and since the appearance of each cell depends on its stage in the cell cycle, standard image processing techniques cannot be used for localization. This paper addresses these challenges by segmenting the spatio-temporal volume using level sets. Experimental results show the promise of this approach

Masked Object Registration in the Fourier Domain

Registration is one of the most common tasks of image analysis and computer vision applications. The requirements of most registration algorithms include large capture range and fast computation so that the algorithms are robust to different scenarios and can be computed in a reasonable amount of time. For these purposes, registration in the Fourier domain using normalized cross-correlation is well suited and has been extensively studied in the literature. Another common requirement is masking, which is necessary for applications where certain regions of the image that would adversely affect the registration result should be ignored. To address these requirements, we have derived a mathematical model that describes an exact form for embedding the masking step fully into the Fourier domain so that all steps of translation registration can be computed efficiently using Fast Fourier Transforms. We provide algorithms and implementation details that demonstrate the correctness of our derivations. We also demonstrate how this masked FFT registration approach can be applied to improve the Fourier-Mellin algorithm that calculates translation, rotation, and scale in the Fourier domain. We demonstrate the computational efficiency, advantages, and correctness of our algorithm on a number of images from real-world applications. Our framework enables fast, global, parameter-free registration of images with masked regions.

Bias in the Localization of Curved Edges

This paper presents a theoretical and experimental analysis of the bias in the localization of edges detected from the zeros of the second derivative of the image in the direction of its gradient, such as the Canny edge detector. Its contributions over previous art are: a quantification of the localization bias as a function of the scale σ of the smoothing filter and the radius of curvature R of the edge, which unifies, without any approximation, previous results that independently studied the case of R≫σ or σ≫ R; the determination of an optimal scale at which edge curvature can be accurately recovered for circular objects; and a technique to compensate for the localization bias which can be easily incorporated into existing algorithms for edge detection. The theoretical results are validated by experiments with synthetic data, and the bias correction algorithm introduced here is reduced to practice on real images.

Masked FFT registration

Registration is a ubiquitous task for image analysis applications. Generally, the requirements of registration algorithms include fast computation and large capture range. For these purposes, registration in the Fourier domain using normalized cross correlation is well suited and has been extensively studied in the literature. Another common requirement is masking, which is necessary for applications where certain regions of the image that would adversely affect the registration result should be ignored. To address these requirements, we have derived a mathematical model that describes an exact form for embedding the masking step fully into the Fourier domain. We also provide an extension of this masked registration approach from simple translation to also include rotation and scale. We demonstrate the computational efficiency of our algorithm and validate its correctness on several synthetic images and real ultrasound images. Our framework enables fast, global, parameter-free registration of images with masked regions.

Tracking of Carotid Arteries in Ultrasound Images

We introduce an automated method for the 3D tracking of carotids acquired as a sequence of 2D ultrasound images. The method includes an image stabilization step that compensates for the cardiac and respiratory motion of the carotid, and tracks the carotid wall via a shape and appearance model trained from representative images. Envisaging an application in automatic detection of plaques, the algorithm was tested on ultrasound volumes from 4,000 patients and its accuracy was evaluated by measuring the distance between the location of more than 4,000 carotid plaques and the location of the carotid wall as estimated by the proposed algorithm. Results show that the centroids of over 95% of the carotid plaques in the dataset were located within 3 mm of the estimated carotid wall, indicating the accuracy of the tracking algorithm.