No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Segmentation of complex cell clusters in microscopic images: Application to bone marrow samples
Abstract
Morphologic examination of bone marrow and peripheral blood samples continues to be the cornerstone in diagnostic hematology. In recent years, interest in automatic leukocyte classification using image analysis has increased rapidly. Such systems collect a series of images in which each cell must be segmented accurately to be classified correctly. Although segmentation algorithms have been developed for sparse cells in peripheral blood, the problem of segmenting the complex cell clusters characterizing bone marrow images is harder and has not been addressed previously.
We present a novel algorithm for segmenting clusters of any number of densely packed cells. The algorithm first oversegments the image into cell subparts. These parts are then assembled into complete cells by solving a combinatorial optimization problem in an efficient way.
Our experimental results show that the algorithm succeeds in correctly segmenting densely clustered leukocytes in bone marrow images.
The presented algorithm enables image analysis-based analysis of bone marrow samples for the first time and may also be adopted for other digital cytometric applications where separation of complex cell clusters is required.
... Some variations of these algorithms are also adapted to solve the white blood cell segmentation problem. These methods usually employ distance transformations (36,37) for marker detection and use gradients/intensities for marking function definition (38). However, they usually yield oversegmented cells and/or irregular cell boundaries, which are refined afterwards (37,39). ...
... These methods usually employ distance transformations (36,37) for marker detection and use gradients/intensities for marking function definition (38). However, they usually yield oversegmented cells and/or irregular cell boundaries, which are refined afterwards (37,39). These methods address the problem of segmenting leukemic cells to an extent (36)(37)(38), but there still remain challenges to overcome when predefined markers do not represent cells accurately. ...
... However, they usually yield oversegmented cells and/or irregular cell boundaries, which are refined afterwards (37,39). These methods address the problem of segmenting leukemic cells to an extent (36)(37)(38), but there still remain challenges to overcome when predefined markers do not represent cells accurately. For white blood cell segmentation, this problem arises when an image contains highly over-layered leukemic cells with fuzzy boundaries and aggravates if the image also contains confluent red blood cells adjacent to white blood ones. ...
Computer-based imaging systems are becoming important tools for quantitative assessment of peripheral blood and bone marrow samples to help experts diagnose blood disorders such as acute leukemia. These systems generally initiate a segmentation stage where white blood cells are separated from the background and other nonsalient objects. As the success of such imaging systems mainly depends on the accuracy of this stage, studies attach great importance for developing accurate segmentation algorithms. Although previous studies give promising results for segmentation of sparsely distributed normal white blood cells, only a few of them focus on segmenting touching and overlapping cell clusters, which is usually the case when leukemic cells are present. In this article, we present a new algorithm for segmentation of both normal and leukemic cells in peripheral blood and bone marrow images. In this algorithm, we propose to model color and shape characteristics of white blood cells by defining two transformations and introduce an efficient use of these transformations in a marker-controlled watershed algorithm. Particularly, these domain specific characteristics are used to identify markers and define the marking function of the watershed algorithm as well as to eliminate false white blood cells in a postprocessing step. Working on 650 white blood cells in peripheral blood and bone marrow images, our experiments reveal that the proposed algorithm improves the segmentation performance compared with its counterparts, leading to high accuracies for both sparsely distributed normal white blood cells and dense leukemic cell clusters. © 2014 International Society for Advancement of Cytometry
... One of the major complications in the automatic segmentation of cellular images arises due to the fact that the cells are often closely clustered. Cluster division to isolate individual cell objects is a critical task in the automatic evaluation of cytological and histological images to study the morphology of cells or nuclei and/or the spatial distribution of cells in the tissue specimen (Malpica et al., 1997; Pal et al., 1998; Nilsson & Heyden, 2005; Schmitt & Hasse, 2009). Several automatic segmentation methods such as thresholding, simple region growing, edge or boundary detection methods have been reported in the literature for cell or nuclei segmentation. ...
... However, they cannot be applied to isolate cells from the cluster. To address this issue, significant attempts have been made specifically towards the automatic cluster splitting for cells or cell nuclei image (Vincent & Soille, 1991; Yeo et al., 1993; Ancin et al., 1995 Ancin et al., , 1996 Fernandez et al., 1995; Najman & Schmitt, 1996; Malpica et al., 1997; Wang, 1998; Solorzano et al., 1999; Adiga & Chaudhuri, 2001; Kumar et al., 2002 Kumar et al., , 2006 Ruberto et al., 2002; Lin et al., 2003; Wählby et al., 2003; Wählby et al., 2004; Nilsson & Heyden, 2005; Gniadek & Warren, 2007; Li et al., 2007; Long et al., 2007; Bai et al., 2009; Schmitt & Hasse, 2009; Yu et al., 2009; Schmitt & Reetz, 2009; Zhong et al., 2009). Watershed segmentation (Vincent & Soille, 1991) is one of the most valuable and popular tools. ...
... To address this issue, many improved watershed methods have been proposed. One of the wellknown solutions is marker controlled or seeded watershed transformation (Ancin et al., 1995Ancin et al., , 1996 Malpica et al., 1997; Solorzano et al., 1999; Ruberto et al., 2002; Lin et al., 2003; Wählby et al., 2003 Nilsson & Heyden, 2005; Gniadek & Warren, 2007; Schmitt & Hasse, 2009). In these methods, watershed transformation is performed by defining a singular seed or marker for each cell or nuclei object using grey scale mathematical morphological operations based on both morphological and intensity information. ...
With the rapid advance of three-dimensional (3D) confocal imaging technology, more and more 3D cellular images will be available. Segmentation of intact cells is a critical task in automated image analysis and quantification of cellular microscopic images. One of the major complications in the automatic segmentation of cellular images arises due to the fact that cells are often closely clustered. Several algorithms are proposed for segmenting cell clusters but most of them are 2D based. In other words, these algorithms are designed to segment 2D cell clusters from a single image. Given 2D segmentation methods developed, they can certainly be applied to each image slice with the 3D cellular volume to obtain the segmented cell clusters. Apparently, in such case, the 3D depth information with the volumetric images is not really used. Often, 3D reconstruction is conducted after the individualized segmentation to build the 3D cellular models from segmented 2D cellular contours. Such 2D native process is not appropriate as stacking of individually segmented 2D cells or nuclei do not necessarily form the correct and complete 3D cells or nuclei in 3D. This paper proposes a novel and efficient 3D cluster splitting algorithm based on concavity analysis and interslice spatial coherence. We have taken the advantage of using the 3D boundary points detected using higher order statistics as an input contour for performing the 3D cluster splitting algorithm. The idea is to separate the touching or overlapping cells or nuclei in a 3D native way. Experimental results show the efficiency of our algorithm for 3D microscopic cellular images.
... One of the major complications in the automatic segmentation of cellular images arises due to the fact that the cells are often closely clustered. Cluster division to isolate individual cell objects is a critical task in the automatic evaluation of cytological and histological images to study the morphology of cells or nuclei and/or the spatial distribution of cells in the tissue specimen (Malpica et al., 1997;Pal et al., 1998;Nilsson & Heyden, 2005;Schmitt & Hasse, 2009). ...
... However, they cannot be applied to isolate cells from the cluster. To address this issue, significant attempts have been made specifically towards the automatic cluster splitting for cells or cell nuclei image (Vincent & Soille, 1991;Yeo et al., 1993;Ancin et al., 1995Ancin et al., , 1996Fernandez et al., 1995;Najman & Schmitt, 1996;Malpica et al., 1997;Wang, 1998;Solorzano et al., 1999;Adiga & Chaudhuri, 2001;Kumar et al., 2002Kumar et al., , 2006Ruberto et al., 2002;Lin et al., 2003;Wählby et al., 2003;Wählby et al., 2004;Nilsson & Heyden, 2005;Gniadek & Warren, 2007;Li et al., 2007;Long et al., 2007;Bai et al., 2009;Schmitt & Hasse, 2009;Yu et al., 2009;Schmitt & Reetz, 2009;Zhong et al., 2009). ...
... To address this issue, many improved watershed methods have been proposed. One of the wellknown solutions is marker controlled or seeded watershed transformation (Ancin et al., 1995(Ancin et al., , 1996Malpica et al., 1997;Solorzano et al., 1999;Ruberto et al., 2002;Lin et al., 2003;Wählby et al., 2003Wählby et al., , 2004Nilsson & Heyden, 2005;Gniadek & Warren, 2007;Schmitt & Hasse, 2009). In these methods, watershed transformation is performed by defining a singular seed or marker for each cell or nuclei object using grey scale mathematical morphological operations based on both morphological and intensity information. ...
With the rapid advance of three-dimensional (3D) confocal imaging technology, more and more 3D cellular images will be available. Segmentation of intact cells is a critical task in automated image analysis and quantification of cellular microscopic images. One of the major complications in the automatic segmentation of cellular images arises due to the fact that cells are often closely clustered. Several algorithms are proposed for segmenting cell clusters but most of them are 2D based. In other words, these algorithms are designed to segment 2D cell clusters from a single image. Given 2D segmentation methods developed, they can certainly be applied to each image slice with the 3D cellular volume to obtain the segmented cell clusters. Apparently, in such case, the 3D depth information with the volumetric images is not really used. Often, 3D reconstruction is conducted after the individualized segmentation to build the 3D cellular models from segmented 2D cellular contours. Such 2D native process is not appropriate as stacking of individually segmented 2D cells or nuclei do not necessarily form the correct and complete 3D cells or nuclei in 3D. This paper proposes a novel and efficient 3D cluster splitting algorithm based on concavity analysis and interslice spatial coherence. We have taken the advantage of using the 3D boundary points detected using higher order statistics as an input contour for performing the 3D cluster splitting algorithm. The idea is to separate the touching or overlapping cells or nuclei in a 3D native way. Experimental results show the efficiency of our algorithm for 3D microscopic cellular images.
... Edge-based approaches extract edges of the cell nucleus and plasma to segment the image using for example the Canny edge detector or the Teager/Kaiser filter [5]. Contour-based methods use active-contour methods (snakes) or level-set approaches to extract plasma and nucleus contours [8], [9]. Pixel-based approaches are strongly represented in current literature since color features seem to be a very efficient criterion for separating the cell regions due to the fact that blood smears are stained prior to image acquisition. ...
... We propose a level-set based approach in order to refine the leukocyte contour. The level-set method for image segmentation has been described by Sethian [12] and has already been used for leukocyte segmentation by Nilsson et al. [8]. The level-set approach defines a contour v = {(x, y)|Φ(x, y) = 0} as zero level-set of a helper function Φ. ...
... α and β are parameters which define the influence of curvature and the vector-field on the contour evolution. As curvature measure we use the approximation provided by Nilsson et al. [8]. Usually G is defined in relation to the image gradient ∇I . ...
Differential blood count is a standard method in hematological laboratory diagnosis. In the course of developing a computer-assisted microscopy system for the generation of differential blood counts, the detection and segmentation of white and red blood cells forms an essential step and its exactness is a fundamental prerequisite for the effectiveness of the subsequent classification step. We propose a method for the exact segmentation of leukocytes and erythrocytes in a simultaneous and cooperative way. We combine pixel-wise classification with template matching to locate erythrocytes and use a level-set approach in order to get the exact cell contours of leukocyte nucleus and plasma regions as well as erythrocyte regions. An evaluation comparing the performance of the algorithm to the manual segmentation performed by several persons yielded good results.
... The separation of connected cells is still a great challenge. Several papers provide different approaches to separate cells of a specific type [8][9][10][11]. Due to the often simple morphology of the analyzed cell types, a separation of clusters with simple rules is possible. ...
... In comparison, the method described in [9] results in 3.9% deviation from the reference cell count for leukocyte cells. In contrast to type L929, the leukocyte cells morphology does not vary so much. ...
This paper presents a new method to separate cells on microscopic surfaces joined together in cell clusters into individual cells. Important features of this method are that the remaining object geometry is preserved and few contour points are required for finding joints between neighboring cells. There are alternative methods such as morphological operations or the watershed transformation based on the inverse distance transformation but they have certain disadvantages compared to the method presented in this paper. The discussed method contains knowledge-based components in form of a decision function and exchangeable rules to avoid unwanted separations.
... In particular, the separation of agglomerated cell areas has often been ignored in other publications or is only suitable for a type of cells with specific geometric characteristics (Kothari et al., 2009;Nilsson and Heyden, 2005;Sheehy et al., 2008;Weixing and Hao, 2007). ...
... In contrast to publications that depend on geometrical characteristics (Buhl et al., 2010;Kothari et al., 2009;Metzler et al., 1999;Nilsson and Heyden, 2005;Sheehy et al., 2008;Weixing and Hao, 2007) or the detection of colour differences for the cell separation, the method described in this paper is entirely independent from the morphology of the cells to be detected. It is thus also suitable for the separation of cell types that form compact agglomerates, for example MC3T3 type cells. ...
This paper presents a method of separating cells that are connected to each other forming clusters. The difference to many other publications covering similar topics is that the cell types we are dealing with form clusters of highly varying morphology. An advantage of our method is that it can be universally used for different cell types. The segmentation method is based on a growth simulation starting from the nuclei areas. To start the evaluation, the cells need to be made visible with a histological stain, in our case with the May-Grünwald solution. After the staining process has been completed, the nuclei areas can be distinguished from the other cell areas by a histogram backprojection algorithm. The presented method can, in addition to histological stained cells, also be applied to fluorescent-stained cells.
... Staining and illumination inconsistences can lead to gray value variety, which is a big trouble for image segmentation. (3) The big variability of the white blood cells characteristics. The cells are frequently clustered, and there is no clear boundary between the nucleus and cytoplasm in many cases. ...
... In the past decades, a lot of meaningful work based on color images was done in bone marrow images segmentation [2][3][4][5][6][7]. But for standard color image analysis, there are many problems [8]: Firstly, in image acquisition process, the images are illuminantdependent and the image quality is influenced by spectral characteristics of the imaging system, such as spectral responses of lens, optic ununiformity and throughput properties, which makes image reproduction very difficult. ...
Counting of different classes of white blood cells in bone marrow smears can give pathologists valuable information regarding various hematological disorders. For automation imaging analysis techniques, precise segmentation of White Blood Cells is quite challenging due to the complex contents in bone marrow smears. Far more different from traditional color imaging analysis methods, we introduced multispectral imaging techniques. After a high quality image was acquired, the spectrum of each pixel was directly fed into a trained Support Vector Machine (SVM) for classification, and then morphological binary operations were performed to correct the small error-classified regions. Mass of experiments showed that the segmentation results are highly satisfactory and inspiring. It shows that the introduction of multispectral imaging analysis techniques into White Blood Cells detection is a success. Multispectral imaging analysis is a promising technique in biomedicine.
... To speed up the general level set framework, many approximations have been introduced in recent years. In particular, the narrow-band algorithm [10] used for segmentation of nuclei and cells labeled with the fluorescently labeled membranes [17] or the fast marching method [18] used for segmentation of cell clusters in two-dimensional bone marrow samples [19]. The fast marching method was also used for 3D reconstruction of interphase chromosomes [20,21]. ...
... The third recently published algorithm was proposed by Nilsson and Heyden [3]. Although this method and its later usage in biomedical image processing [19] were presented only for 2D data, its extension to three dimensions is straightforward. Like in the DTA, the evolution of an initial interface is realized by a point-by-point manner. ...
Image segmentation, one of the fundamental task of image processing, can be accurately solved using the level set framework.
However, the computational time demands of the level set methods make them practically useless, especially for segmentation
of large three-dimensional images. Many approximations have been introduced in recent years to speed up the computation of
the level set methods. Although these algorithms provide favourable results, most of them were not properly tested against
ground truth images. In this paper we present a comparison of three methods: the Sparse-Field method[1], Deng and Tsui’s
algorithm[2] and Nilsson and Heyden’s algorithm[3]. Our main motivation was to compare these methods on 3D image data acquired
using fluorescence microscope, but we suppose that presented results are also valid and applicable to other biomedical images
like CT scans, MRI or ultrasound images. We focus on a comparison of the method accuracy, speed and ability to detect several
objects located close to each other for both 2D and 3D images. Furthermore, since the input data of our experiments are artificially
generated, we are able to compare obtained segmentation results with ground truth images.
... In the past decades, a lot of meaningful work had been done in bone marrow images segmentation [2][3][4][5][6][7][8][9][10][11]. Most of their work mainly focused on two difficult points: the segmentation of nuclei and the cytoplasm, along with cluster segmentation. ...
... Distance transform (i.e. a function associating to every pixel of a binary image its distance to the border) of the segmented clumped shape is also used in some clump splitting approaches. The maxima in the distance image can be used as markers for subsequent segmentation of the original image by the watershed algorithm (Malpica et al. 1997;Angulo & Flandrin, 2003a, Nilsson & Heyden, 2005 or another region growing approach (Hengen et al., 2002). But the watershed segmentation can also be applied directly on the distance image for separating circular shapes (Lindbland, 2002). ...
... Recently, convolutional neural networks were applied to leukocyte counting [8]. Mature blood cell recognition from peripheral blood smears has been done in [6], and others worked on bone marrow material classifying pathological megakaryocytes [9], leukocytes [10], or myelocytes from myelogenous leukemia [3]. However, the quantification of blood cell maturation in the bone marrow has not been sufficiently studied yet. ...
Classification of cell types in context of the architecture in tissue specimen is the basis of diagnostic pathology and decisions for comprehensive investigations rely on a valid interpretation of tissue morphology. Especially visual examination of bone marrow cells takes a considerable amount of time and inter-observer variability can be remarkable. In this work, we propose a novel rotation-invariant learning scheme for multi-class Echo State Networks (ESNs), which achieves very high performance in automated bone marrow cell classification. Based on repre- senting static images as temporal sequence of rotations, we show how ESNs robustly recognize cells of arbitrary rotations by taking advantage of their short-term memory capacity.
... 2. The lack of contrast in the neurosphere allowing to detect the cells membrane in the neurosphere. concentrate on segmenting the clusters, by then proceeding to a more precise segmentation of the cluster, using either active contours [75,19,110] or graph-cuts [23]. Based on ellipse fitting approach for segmenting simple cluster cells [48], we propose to segment the cells contained into the neurosphere, using a circle fitting process. ...
The study of stem cells is one of the most important fields of research in the biomedical field. Computer vision and image processing have been greatly emphasized in this area for the development of automated solutions for culture and observation of cells. This work proposes a new methodology for observing and modelling the proliferation of neural stem cell under a phase contrast microscope. At each time lapse observation performed by the microscope during the proliferation, the system determines a three-dimensional model of the structure formed by the observed cells. This is achieved by a framework combining analysis, synthesis and selection process. First, an analysis of the images from the microscope segments the neurosphere and the constituent cells. With this analysis, combined with prior knowledge about the cells and their culture protocol, several 3-D possible models are generated through a synthesis process. These models are finally selected and evaluated according to their likelihood with the microscope image using a 3-D to 2-D registration method. Through this approach, we present an automatic visualisation tool and observation of the proliferation of neural stem cell under a phase contrast microscope.
... In the last phase leukocyte cells are separated by using Fast Marching methods. Nilsson et al. [2,3] describe a method for the segmentation of complex cell clusters. At first a background/foreground separation is done by thresholding. ...
For the diagnosis of leukemia the morphological analysis of bone marrow is essential. This procedure is time consuming, partially subjective, error-prone and cumbersome. Moreover, repeated examinations may lead to intra- and inter-observer variances. Therefore, an automation of the bone marrow analysis is pursued. The automatic classification of bone marrow cells is highly dependent on the preceding segmentation of the nucleus and plasma parts of the cell. In this contribution we propose a dynamic programming approach for the segmentation of already localized bone marrow cells and evaluate the method with 1000 manually segmented cells. With this approach the segmentation quality for whole cells is 0.93 and 0.85 for the corresponding nucleus parts.
... Another image analysis challenge involves the segmentation of tightly packed cells, which is necessary for automated quantification of many cell behaviors. Previous studies addressed this challenge with the help of fluorescent nuclear or membrane markers [see (Doncic, Eser, Atay, & Skotheim, 2013;Indhumathi, Cai, Guan, & Opas, 2011;Malpica et al., 1997;Nilsson & Heyden, 2005;Schmitt & Reetz, 2009) and references therein]. However, cell behaviors of interest must frequently be detected at the same time using fluorescent probes (e.g. to track intracellular localization of proteins or movements of subcellular structures); and additional markers for segmentation may obscure the behavior of interest. ...
Understanding the heterogeneous dynamics of cellular processes requires not only tools to visualize molecular behavior but also versatile approaches to extract and analyze the information contained in live-cell movies of many cells. Automated identification and tracking of cellular features enable thorough and consistent comparative analyses in a high-throughput manner. Here, we present tools for two challenging problems in computational image analysis: (1) classification of motion for cells with complex shapes and dynamics and (2) segmentation of clustered cells and quantification of intracellular protein distributions based on a single fluorescence channel. We describe these methods and user-friendly software(1) (MATLAB applications with graphical user interfaces) so these tools can be readily applied without an extensive knowledge of computational techniques.
... Das beschriebene Verfahren zur automatischen Bestimmung der Zellzentren in Knochenmarkbildern liefert sehrüberzeugende Ergebnisse, sodass mehrere Einsatzzwecke möglich sind. Zum Einen lassen sich beispielsweise Segmentierungsverfahren für Knochenmarkzellen, wie sie etwa in [2,3,4] ...
Die morphologische Analyse von Knochenmarkpräparaten ist bedeutend für die Leukämiediagnose. Bisher wird dabei das Ausz ählen und Klassifizieren der unterschiedlichen Knochenmarkzellen manuell unter dem Mikroskop durchgeführt und ist zeitaufwändig, z.T. subjektiv und mühsam. Aus diesem Grund wird eine Automatisierung der Analyse von Knochenmarkpräparaten angestrebt. Die automatische Lokalisierung der Zellen stellt dabei die Basis für die nachfolgenden Verarbeitungsschritte, d.h. für die Segmentierung und automatische Klassifikation, dar. Das entwickelte Verfahren löst diese Aufgabe durch zwei unterschiedliche Ansätze für Bilder mit einem niedrigen und einem hohen Zellanteil. Das vorgestellte Verfahren wird mit 400 Knochenmarkbildern aus 200 unterschiedlichen Präparaten evaluiert. Für diese Bilder ergibt sich für die Detektion eine durchschnittliche Sensitivität von 97% bei einer mittleren Falschdetektionsrate von 8%.
... Nowadays, the rapid progress of information technology promotes the automatization of blood cells diagnosis based on modern image processing and pattern recognition approaches [10,13]. Among these techniques, automatic segmentation of certain blood cells out of complex scenes should be foremost solved in quantitative analysis of blood cells. ...
A molecular spectral imaging system instead of common microscope was used to capture the spectral images of blood smears. Then an improved spectral angle mapper algorithm for automatic blood cells segmentation was presented. In this algorithm, the spectral vectors of blood cells were normalized first. Then the spectral angles of all bands and partial bands were calculated respectively. Finally, the blood cells were segmented according to the spectral angles combined with the threshold segmentation method. As a case study, the leukemia cells were selected as the target and segmented with the new algorithm. The results demonstrate that this algorithm can utilizes both spectral and spatial information of blood cells and segment the leukemia cells more accurately.
... In recent years, many automated blood cell morphological analysis systems based on imaging cytometry (5)(6)(7)(8)(9)(10)(11) or imaging flow cytometry (12,13) have been developed. These systems concentrate on the automatic differentiation of leukocytes and provide many informative hematological parameters, intending to replace the manual microscopic review of blood smears. ...
Abnormal neutrophil nucleus lobation (like left shift and right shift) helps to diagnose for some clinical conditions. Currently, quantification of it depends on the manual microscopic inspection of blood smears by clinicians. The quality of the manual inspection is extremely limited by the efficiency of clinicians and their medical background. This article proposed an automatic lobe counting method based on the graph representation of the nucleus region skeletons. Skeletons of the segmented nucleus regions are computed by augmented Fast Marching Method and transformed into plane graphs. Then the nucleus lobes are split based on the extracted graph properties including width distribution along the skeleton and graph structure decomposition. Experiments show that the proposed method could efficiently approaches the real lobe numbers in blood smears and reliably distinguish the stabs from the segmented neutrophils, thus it should be helpful in automatic neutrophil lobe number quantification and abnormal lobation diagnosis.
... Many image processing algorithms have been proposed for cell segmenta- tion, however segmentation of cell populations which grow in complex clusters is still a challenging issue [6]. This paper deals with individual nucleus mod- elling and segmentation of cell clusters from fluorescence labelled images. ...
This paper deals with individual nucleus modelling and segmentation, from fluorescence labelled images, of cell populations
growing in complex clusters. The proposed approach is based on models and operators from mathematical morphology. Cells are
individually marked by the ultimate opening and then are segmented by the watershed transformation. A cell counting algorithm
based on classical results of Boolean model theory is heuristically used to detect errors in segmenting clustered nuclei.
... To increase the classification accuracy especially for the immature cells, such an exact segmentation is essential. Most of the state of the art algorithms [1][2][3][4][5][6][7][8][9][10][11][12] are not able to cope with the challenges related to segmentation of all the different cell classes. The problem of detecting and segmenting nuclei shadows has not been concerned in literature at all. ...
The exact segmentation of nucleus and plasma of a white blood cell (leukocyte) is the basis for the creation of an automatic, image based differential white blood cell count(WBC). In this contribution we present an approach for the according segmentation of leukocytes. For a valid classification of the different cell classes, a precise segmentation is essential. Especially concerning immature cells, which can be distinguished from their mature counterparts only by small differences in some features, a segmentation of nucleus and plasma has to be as precise as possible, to extract those differences. Also the problems with adjacent erythrocyte cells and the usage of a LED illumination are considered. The presented approach can be separated into several steps. After preprocessing by a Kuwahara-filter, the cell is localized by a simple thresholding operation, afterwards a fast-marching method for the localization of a rough cell boundary is defined. To retrieve the cell area a shortest-path-algorithm is applied next. The cell boundary found by the fast-marching approach is finally enhanced by a post-processing step. The concluding segmentation of the cell nucleus is done by a threshold operation. An evaluation of the presented method was done on a representative sample set of 80 images recorded with LED illumination and a 63-fold magnification dry objective. The automatically segmented cell images are compared to a manual segmentation of the same dataset using the Dice-coefficient as well as Hausdorff-distance. The results show that our approach is able to handle the different cell classes and that it improves the segmentation quality significantly.
... A range of approaches for evaluating biomedical segmentation algorithms appear in the literature. Visual inspection of the segmentation algorithm results is perhaps the simplest and most intuitive; it is also widely used (21)(22)(23)(24)(25)(26)(27). Another approach is the use of a metric which quantitatively assesses the performance against reference segmentations. ...
The analysis of fluorescence microscopy of cells often requires the determination of cell edges. This is typically done using segmentation techniques that separate the cell objects in an image from the surrounding background. This study compares segmentation results from nine different segmentation techniques applied to two different cell lines and five different sets of imaging conditions. Significant variability in the results of segmentation was observed that was due solely to differences in imaging conditions or applications of different algorithms. We quantified and compared the results with a novel bivariate similarity index metric that evaluates the degree of underestimating or overestimating a cell object. The results show that commonly used threshold-based segmentation techniques are less accurate than k-means clustering with multiple clusters. Segmentation accuracy varies with imaging conditions that determine the sharpness of cell edges and with geometric features of a cell. Based on this observation, we propose a method that quantifies cell edge character to provide an estimate of how accurately an algorithm will perform. The results of this study will assist the development of criteria for evaluating interlaboratory comparability.
... In consequence, automated accurate detection and segmentation of neurons from microscopic images has been extensively studied (Liu et al., 2008). In general, these algorithms can be divided into three categories: threshold-based (Wu et al., 2000;Wu et al., 1995), watershedbased (Lin et al., 2003;Lin et al., 2005;Malpica et al., 1997;Nilsson and Heyden, 2005;Vincent and Soille, 1991) and model-based approaches (Chang and Parvin, 2006;Li et al., 2006;Lin et al., 2007;Lin et al., 2005;Raman et al., 2007;Ranzato et al., 2007). ...
We present a novel approach for automated detection of neuron somata. A three-step processing pipeline is described on the example of confocal image stacks of NeuN-stained neurons from rat somato-sensory cortex. It results in a set of position landmarks, representing the midpoints of all neuron somata. In the first step, foreground and background pixels are identified, resulting in a binary image. It is based on local thresholding and compensates for imaging and staining artifacts. Once this pre-processing guarantees a standard image quality, clusters of touching neurons are separated in the second step, using a marker-based watershed approach. A model-based algorithm completes the pipeline. It assumes a dominant neuron population with Gaussian distributed volumes within one microscopic field of view. Remaining larger objects are hence split or treated as a second neuron type. A variation of the processing pipeline is presented, showing that our method can also be used for co-localization of neurons in multi-channel images. As an example, we process 2-channel stacks of NeuN-stained somata, labeling all neurons, counterstained with GAD67, labeling GABAergic interneurons, using an adapted pre-processing step for the second channel. The automatically generated landmark sets are compared to manually placed counterparts. A comparison yields that the deviation in landmark position is negligible and that the difference between the numbers of manually and automatically counted neurons is less than 4%. In consequence, this novel approach for neuron counting is a reliable and objective alternative to manual detection.
The article focuses on the concepts of Cell Image Segmentation (CIS) and the gradual introduction of cell counting. Motivated by the rapid development of Machine learning (ML) methods, which is carried out in this investigation. ML is evolving from theory to practical applications, with deep neural network models extensively used in academia and business for various applications, including image counting and natural language processing. These advancements can greatly influence medical imaging technologies, data processing, diagnostics, and healthcare in general. Main objectives of the research are to provide an overview of biological cell counting methods in microscopic images and to explore deep learning (DL)-based image segmentation approaches. The study expertly describes current trends, cutting-edge learning technologies, and platforms utilized for DL approaches. Cell counting is one of the most researched and challenging subjects in computer vision systems. Academics are increasingly interested in this area due to its real-time applications in biology, biochemistry, medical diagnostics, computer vision-based cell tracking systems for large populations, and stem cell manufacturing. Counting cells in the biological field is beneficial. For instance, the ratio of white blood cells to cancer cells in the blood can help determine the origin of a disease. Biologists also need to count cells within cell cultures to monitor the time-dependent growth of cells during bacterial experiments. Numerous methods for cell counting have been developed, after addressing the challenges with Cell Counting algorithms; the article explores promising future directions in CIS and cell counting research fields.
Automotive image segmentation systems are becoming an important tool in the medical field for disease diagnosis. The white blood cell (WBC) segmentation is crucial, because it plays an important role in the determination of the diseases and helps experts to diagnose the blood disease disorders. The precise segmentation of the WBCs is quite challenging because of the complex contents in the bone marrow smears. In this paper, a novel neural network (NN) classifier is proposed for the classification of the bone marrow WBCs. The proposed NN classifier integrates the fractional gravitation search (FGS) algorithm for updating the weight in the radial basis function mapping for the classification of the WBC based on the cell nucleus feature. The experimentation of the proposed FGS-RBNN classifier is carried on the images collected from the publically available dataset. The performance of the proposed methodology is evaluated over the existing classifier approaches using the measures accuracy, sensitivity, and specificity. The results show that the classification using the nucleus features alone can be utilized to achieve the classification with the better accuracy. Moreover, the classification performance of the proposed FGS-RBNN is better than the existing classifiers, and it is proved to be the efficacious classifier with a classification accuracy of 95%.
Estimating the number of blood cells in a sample is an important task in biological research. However, manual counting of cells in microscopy images of counting chambers is very time-consuming. We present an image processing method for detecting the chamber grid and the cells, based on their similarity to an automatically selected sample cell. Due to this approach, the method does not depend on specific cell structure, and can be used for blood cells of different species without adjustments. If deemed appropriate, user interaction is allowed to select the sample cell and adjust the parameters manually. We also present the accuracy and speed evaluation of the method.
Visual examination of blood and bone marrow smears is an important tool for diagnosis, prevention and treatment of clinical patients. The interest of computer aided decision has been identified in many medical applications: automatic methods are being explored to detect, classify and measure objects in hematological cytology. This chapter presents a comprehensive review of the state of the art and currently available literature and techniques related to automated analysis of blood smears. The most relevant image processing and machine learning techniques used to develop a fully automated blood smear analysis system which can help to reduce time spent for slide examination are presented. Advances in each component of this system are described in acquisition, segmentation and detection of cell components, feature extraction and selection approaches for describing the objects, and schemes for cell classification.
Abnormal localization of immature precursors (ALIP) aggregating and clustering in bone marrow biopsy appears earlier than that of bone marrow smears in detection of the relapse of acute myelocytic leukemia (AML). But traditional manual ALIP recognition has many shortcomings such as prone to false alarms, neglect of distribution law before three immature precursor cells gathered, and qualitative analysis instead of quantitative one. So, it is very important to develop a novel automatic method to identify and localize immature precursor cells for computer-aided diagnosis, to disclose their patterns before ALIP with the development of AML. The contributions of this paper are as follows. (1) After preprocessing the image with Otsu method, we identify both precursor cells and trabecular bone by multiple morphological operations and thresholds. (2) We localize the precursors in different regions according to their distances with the nearest trabecular bone based on chamfer distance transform, followed by discussion for the presumptions and limitations of our method. The accuracy of recognition and localization is evaluated based on a comparison with visual evaluation by two blinded observers.
Segmenting and interesting objects from microscopic images and classifying microscopic images are very important for biomedical researching work, which help diagnosis and further biomedical research. However, conventional approaches don't behavior as well as expected when they are applied to solve the problem. We hence propose two methods, radial basis function neural network with fuzzy initialization and graph-based discrete approach, for microscopic image segmenting and classification. The results show that RBF neural network has advantages such as easy to configure and implement, and the training procedure being very fast. In addition, RBF neural network employs fuzzy mean algorithm to accelerate the procedure of parameters and structure selection. Meanwhile, graphed-based discrete approach, which depends on the general formulation of discrete functional regularization on weighted graph, can be used to address cellular extraction segmentation problem.
Computer-aided automatic analysis of microscopic leukocyte is a powerful diagnostic tool in biomedical fields which could reduce the effects of human error, improve the diagnosis accuracy, save manpower and time. However, it is a challenging to segment entire leukocyte populations due to the changing features extracted in the leukocyte image, and this task remains an unsolved issue in blood cell image segmentation. This paper presents an efficient strategy to construct a segmentation model for any leukocyte image using simulated visual attention via learning by on-line sampling. In the sampling stage, two types of visual attention, “bottom-up” and “top-down” together with the movement of the human eye are simulated. We focus on a few regions of interesting and sample high gradient pixels to group training sets. While in the learning stage, the SVM (support vector machine) model is trained in real-time to simulate the visual neuronal system and then classifies pixels and extracts leukocytes from the image. Experimental results show that the proposed method has better performance compared to the marker controlled watershed algorithms with manual intervention and thresholding-based methods.
Early detection of leukemia and reduced risk to human health can result from interdisciplinary integration of image analysis with clinical experimental results. Image analysis relies on efficient and reliable processing algorithms to make quantitative judgments on image data. This article presents the design and implementation of an efficient and high-throughput leukemia cell count and cluster classification algorithm to automatically quantify leukemia population statistics in the field of view. The algorithm is divided into two stages: (1) the cell identification stage and (2) the cell classification and inspection stage. The cell identification stage accurately segments background and noise from foreground pixels. A boundary box is generated enclosing the foreground pixels identifying all isolated cells and cell clusters. The cell classification and inspection stage uses one-dimensional intensity profiles that behave as signature plots to segregate isolated cells from cell clusters and evaluate total count within each cluster. The designed algorithm is tested with a variety of leukemia cell images that vary in image acquisition conditions, image sizes, cell sizes, intensity distributions, and image quality. The proposed algorithm demonstrates good potential in processing both ideal and nonideal images with an average accuracy of 91% and average processing time of 3 s. The performance of the proposed algorithm in comparison to recently published algorithms and commercial image analysis tool further ascertains its robustness.
Analyzing bone marrow is an important task for diagnosing diseases like certain types of leukemia and anemia. There are many different cell types in bone marrow. A certain ratio of these types is characteristic for a healthy human. Each deviation from that ratio is a significant indicator for diseases. Until now determining the ratio is done manually by an expert by counting and classifying the cells with a microscope. This is cumbersome and very time consuming. So there are efforts to automatize the cell counting. The most difficult step to achieve that is the automatic segmentation of leukocytes in bone marrow smears. Because the segmentation quality of existing algorithms is not good enough a new algorithm was developed in the scope of this paper. This new algorithm is robust concerning variations as color fluctuations in the bone marrow images. The evaluation of this algorithm was done by comparing the segmentation results with the results obtained by an existing algorithm. Therefore a set of 27 bone marrow images was segmented and compared against a manual annotation. The segmentation quality obtained by the state of the art algorithm was 0.4544 and the quality achieved by the novel algorithm was 0.645 on a scale from zero to one, zero representing only invalid segmentations and one representing only perfect segmentations.
Reliable cell nuclei segmentation is an important yet unresolved problem in biological imaging studies. This paper presents a novel computerized method for robust cell nuclei segmentation based on gradient flow tracking. This method is composed of three key steps: (1) generate a diffused gradient vector flow field; (2) perform a gradient flow tracking procedure to attract points to the basin of a sink; and (3) separate the image into small regions, each containing one nucleus and nearby peripheral background, and perform local adaptive thresholding in each small region to extract the cell nucleus from the background. To show the generality of the proposed method, we report the validation and experimental results using microscopic image data sets from three research labs, with both over-segmentation and under-segmentation rates below 3%. In particular, this method is able to segment closely juxtaposed or clustered cell nuclei, with high sensitivity and specificity in different situations.
Automated image-based and biochemical assays have greatly increased throughput for quantifying cell numbers in in vitro studies. However, it has been more difficult to automate the counting of specific cell types with complex morphologies in mixed cell cultures. We have developed a fully automated, fast, accurate and objective method for the quantification of primary human GFAP-positive astrocytes and CD45-positive microglia from images of mixed cell populations. This method, called the complex cell count (CCC) assay, utilizes a combination of image processing and analysis operations from MetaMorph (Version 6.2.6, Molecular Devices). The CCC assay consists of four main aspects: image processing with a unique combination of morphology filters; digital thresholding; integrated morphometry analysis; and a configuration of object standards. The time needed to analyze each image is 1.82s. Significant correlations have been consistently achieved between the data obtained from CCC analysis and manual cell counts. This assay can quickly and accurately quantify the number of human astrocytes and microglia in mixed cell culture and can be applied to quantifying a range of other cells/objects with complex morphology in neuroscience research.
Automated segmentation and morphometry of fluorescently labeled cell nuclei in batches of 3D confocal stacks is essential for quantitative studies. Model-based segmentation algorithms are attractive due to their robustness. Previous methods incorporated a single nuclear model. This is a limitation for tissues containing multiple cell types with different nuclear features. Improved segmentation for such tissues requires algorithms that permit multiple models to be used simultaneously. This requires a tight integration of classification and segmentation algorithms. Two or more nuclear models are constructed semiautomatically from user-provided training examples. Starting with an initial over-segmentation produced by a gradient-weighted watershed algorithm, a hierarchical fragment merging tree rooted at each object is built. Linear discriminant analysis is used to classify each candidate using multiple object models. On the basis of the selected class, a Bayesian score is computed. Fragment merging decisions are made by comparing the score with that of other candidates, and the scores of constituent fragments of each candidate. The overall segmentation accuracy was 93.7% and classification accuracy was 93.5%, respectively, on a diverse collection of images drawn from five different regions of the rat brain. The multi-model method was found to achieve high accuracy on nuclear segmentation and classification by correctly resolving ambiguities in clustered regions containing heterogeneous cell populations.
Image cytometry technology has been extended to 3D based on high-speed multiphoton microscopy. This technique allows in situ study of tissue specimens preserving important cell-cell and cell-extracellular matrix interactions. The imaging system was based on high-speed multiphoton microscopy (HSMPM) for 3D deep tissue imaging with minimal photodamage. Using appropriate fluorescent labels and a specimen translation stage, we could quantify cellular and biochemical states of tissues in a high throughput manner. This approach could assay tissue structures with subcellular resolution down to a few hundred micrometers deep. Its throughput could be quantified by the rate of volume imaging: 1.45 mm(3)/h with high resolution. For a tissue containing tightly packed, stratified cellular layers, this rate corresponded to sampling about 200 cells/s. We characterized the performance of 3D tissue cytometer by quantifying rare cell populations in 2D and 3D specimens in vitro. The measured population ratios, which were obtained by image analysis, agreed well with the expected ratios down to the ratio of 1/10(5). This technology was also applied to the detection of rare skin structures based on endogenous fluorophores. Sebaceous glands and a cell cluster at the base of a hair follicle were identified. Finally, the 3D tissue cytometer was applied to detect rare cells that had undergone homologous mitotic recombination in a novel transgenic mouse model, where recombination events could result in the expression of enhanced yellow fluorescent protein in the cells. 3D tissue cytometry based on HSMPM demonstrated its screening capability with high sensitivity and showed the possibility of studying cellular and biochemical states in tissues in situ. This technique will significantly expand the scope of cytometric studies to the biomedical problems where spatial and chemical relationships between cells and their tissue environments are important.
Quantification of cells is a critical step towards the assessment of cell fate in neurological disease or developmental models. Here, we present a novel cell detection method for the automatic quantification of zebrafish neuronal cells, including primary motor neurons, Rohon-Beard neurons, and retinal cells. Our method consists of four steps. First, a diffused gradient vector field is produced. Subsequently, the orientations and magnitude information of diffused gradients are accumulated, and a response image is computed. In the third step, we perform non-maximum suppression on the response image and identify the detection candidates. In the fourth and final step the detected objects are grouped into clusters based on their color information. Using five different datasets depicting zebrafish cells, we show that our method consistently displays high sensitivity and specificity of over 95%. Our results demonstrate the general applicability of this method to different data samples, including nuclear staining, immunohistochemistry, and cell death detection.
This paper presents a high throughput cell count and cluster classification algorithm to quantify population statistics of leukemia cell lines on a conventional hemocytometer. The algorithm has been designed, implemented and tested on test images that vary in image quality. The proposed algorithm uses a recursively segmented, median filtered and a boosted Prewitt gradient mask to generate a boundary box that encloses all the identified cells. Intensity profile plots acting as signature plots further assist in classifying a single isolated cell from a cell cluster. Processed results compared manually by a biological expert resulted in an accuracy of 95 % for even low quality images with a computational time ranging between 8-12sec. Improved performance from the proposed algorithm could be observed when compared with other conventional image analysis tools.
The paper presents a method for automatic localization and feature extraction of white blood cells (WBCs) with color images to develop an efficient automated WBC counting system based on image analysis and recognition. Nucleus blobs extraction consists of five steps: (1) nucleus pixel labeling; (2) filtration of nucleus pixel template; (3) segmentation and extraction of nucleus blobs by region growing; (4) removal of uninterested blobs; and (5) marking of external and internal blob border, and holes pixels. The detection of nucleus pixels is based on the intensity of the G image plane and the balance between G and B intensity. Localized nucleus segments are grouped into a cell nucleus by a hierarchic merging procedure in accordance with their area, shapes and conditions of their spatial occurrence. Cytoplasm segmentation based on the pixel intensity and color parameters is found to be unreliable. We overcome this problem by using an edge improving technique. WBC templates are then calculated and additional cell feature sets are constructed for the recognition. Cell feature sets include description of principal geometric and color properties for each type of WBCs. Finally we evaluate the recognition accuracy of the developed algorithm that is proved to be highly reliable and fast.
The paper presents a method for automatic localization and segmentation of white blood cells (WBCs) with color images to develop an efficient automated leukocyte counter by using pattern recognition-based slide readers. The segmentation techniques consist of the following steps. On the first a smear image acquired at the low magnification. The next is extraction of WBC nuclei by chromatic properties and image mapping. After this the cells clustered according to the distances between them and regions of interest (ROI) determined. Image of ROI captured at the high magnification and its validity checked. Then nucleus segments extracted and grouped into prospective cells. The detection of blood cells is based on the intensity of G image plane and the balance between G and B intensity of the nuclei. A cytoplasm region approximated by a circle area around the nucleus center. Finally, the cytoplasm area cleaned considering a priori knowledge of background color and possible cell occlusions. The result of the segmentation is presented in the form of a cell location list and image template in which every pixel is assigned to a label such as Background, Cytoplasm, Nucleus, Hole, etc. The proposed technique has yielded correct segmentation of complex image scenes for blood smears prepared by ordinary manual staining methods in 99% of tested images.© (1996) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.
Mathematics Version of Record
Human leukocytes (white blood cells) can be divided into about
twenty subclasses and the estimation of their distribution, called
differential counting, is an important diagnostic tool in various
clinical settings. Automatic differential counters based on digital
image analysis require good segmentation algorithms to locate each cell
and the accuracy of the subsequent classification depends on the correct
segmentation of solitary cells as well as cell clusters. Previously
published segmentation algorithms mainly use various thresholding
schemes to extract the nucleus and cytoplasm of solitary cells but, so
far, no successful cluster segmentation method has been developed. In
this paper we present a model-based segmentation algorithm that uses
interface propagation models to locate nuclear segments and their
adherent cytoplasms. These segments are then assembled using a
model-based combinatorial optimization scheme. The results are very
promising and, to our knowledge, this is the first successful attempt to
solve this problem
The objective of this work is to investigate the white blood cell
(WBC) image recognition problem at all stages. A robust and effective
method for automatic WBC differentiation, based on both statistical
pattern recognition and neural net approaches, is presented. We
demonstrate well-evaluated results ranging from image scene segmentation
techniques to recognition details. Recognition accuracy on the test set
of 662 images of five WBC types obtained by different imaging systems
from 22 bloodstains is not less than 98%
A technique for the localization of cytoplasmic and nucleic material in Wrights and Wrights-Giemsa strained blood cells is described. The microscopic field of view is first scanned with three different wavelengths of light and then digitized with the aid of a microscopic/vidicon/computer. Each of the three pictures is then converted to a binary color picture by comparison with three calculated clipping levels. The three binary color pictures are then logically combined to generate ``masks'' or sets of points which correspond to 1)nucleic points in the picture, 2) cytoplasmic points in the picture, and 3)red-cell points in the picture. The algorithms involved are easy to implement either in hardware or software, and execute very rapidly in either environment.
Serial and parallel algorithms for solving a system of equations that arises from the discretization of the Hamilton-Jacobi (HJ) equation associated to a trajectory optimization problem are presented. After discretization, a solution to the system of nonlinear equations whose structure resembles the structure of the original HJ equation is discussed.
For an optical or acoustical wavefront running through a medium of space variant refraction index the eikonal equation connects local front arrival time with local refraction index. So-called difference approximation methods are known for solving the spatial wavefront development with time and thus, indirectly, the eikonal equation.Here a novel fast method for the calculation of an approximative solution of the eikonal equation is proposed.From literature it is known that by solving an eikonal equation one can construct a line pattern rendition of a given image. We have generalized this method and made it fit for line engravings.We have found yet another kind of image display based on solving an eikonal equation: shading from shape. We propose to construct a matte 3-D surface (shape) that, when illuminated perpendicularly and imaged in eye or camera, yields a grey value (shading, luminance) field that renders the image.Both methods have been applied in a recent design for a Dutch coin.
This paper discusses the classification of blood cell images. The image input method using the single CCD color TV camera and the special color compensation filter, as well as the method of region segmentation for the blood cell images, are described. The following elaborations are made in the image input stage to separate the red blood cell and the white blood cell images in a stable way using optical means.
(1) The bluish green light near 450 ∼ 500 nm is cut off from the white light.
(2) The accompanying light imbalance between the blue light and the green/red light is compensated by the special color compensation filter.
In the (region) segmentation of the blood cell images, a logical operation is developed in which the region of the red blood cell is extracted based on the binary images obtained by the threshold processing of the subtracted image. Using those methods, 59 blood cell images are processed and a satisfactory segmentation result is obtained.
A simple technique which automatically detects and then segments nucleated cells in Wright's giemsa-stained blood smears is presented. Our method differs from others in (1) the simplicity of our algorithms; (2) inclusion of touching (as well as nontouching) cells; and (3) use of these algorithms to segment as well as to detect nucleated cells employing conventionally prepared smears. Our method involves: (1) acquisition of spectral images; (2) preprocessing the acquired images; (3) detection of single and touching cells in the scene; (4) segmentation of the cells into nuclear and cytoplasmic regions; and (5) postprocessing of the segmented regions. The first two steps of this algorithm are employed to obtain high-quality images, to remove random noise, and to correct aberration and shading effects. Spectral information of the image is used in step 3 to segment the nucleated cells from the rest of the scene. Using the initial cell masks, nucleated cells which are just touching are detected and separated. Simple features are then extracted and conditions applied such that single nucleated cells are finally selected. In step 4, the intensity variations of the cells are then used to segment the nucleus from the cytoplasm. The success rate in segmenting the nucleated cells is between 81 and 93%. The major errors in segmentation of the nucleus and the cytoplasm in the recognized nucleated cells are 3.5% and 2.2%, respectively. © 1992 Wiley-Liss, Inc.
A major problem in the development of systems for automated differential blood count of leucocytes in the peripheral blood is the correct segmentation of the cell scene, i.e., its decomposition into nucleus, plasma, erythrocytes, thrombocytes, etc. Several algorithms have been proposed in the literature and they differ considerably in performance, expense of implementation, and speed. This paper describes a fast and simple segmentation scheme based on hierarchical thresholding. The thresholds are derived from histograms of the cell scene based on two color features used at different processing steps. Segmentation is achieved by application of several local operators to the cell scene. These operators can easily be implemented in hardware. Therefore the segmentation method is suitable for the implementation on a special fast hardware processor as part of a system for automated differential blood counting.
This paper describes a fast algorithm for topology independent tracking of moving interfaces under curvature- and velocity field-dependent speed laws. This is usually done in the level set framework using the narrow-band algorithm, which accurately solves the level set equation but is too slow to use in real-time or near real-time image segmentation applications. In this paper we introduce a fast algorithm for tracking moving interfaces in a level set-like manner. The algorithm relies on two key components: First, it tracks the interface by scheduling point-wise propagation events using a heap sorted queue. Second, the local geometric properties of the interface are defined so that they can be efficiently updated in an incremental manner and so that they do not require the presence of the signed distance function. Finally examples are given that indicate that the algorithm is fast and accurate enough for near real-time segmentation applications.
A distance transformation converts a binary digital image, consisting of feature and non-feature pixels, into an image where all non-feature pixels have a value corresponding to the distance to the nearest feature pixel. Computing these distances is in principle a global operation. However, global operations are prohibitively costly. Therefore algorithms that consider only small neighborhoods, but still give a reasonable approximation of the Euclidean distance, are necessary. In the first part of this paper optimal distance transformations are developed. Local neighborhoods of sizes up to 7×7 pixels are used. First real-valued distance transformations are considered, and then the best integer approximations of them are computed. A new distance transformation is presented, that is easily computed and has a maximal error of about 2%. In the second part of the paper six different distance transformations, both old and new, are used for a few different applications. These applications show both that the choice of distance transformation is important, and that any of the six transformations may be the right choice.
We devise new numerical algorithms, called PSC algorithms, for following fronts propagating with curvature-dependent speed. The speed may be an arbitrary function of curvature, and the front also can be passively advected by an underlying flow. These algorithms approximate the equations of motion, which resemble Hamilton-Jacobi equations with parabolic right-hand sides, by using techniques from hyperbolic conservation laws. Non-oscillatory schemes of various orders of accuracy are used to solve the equations, providing methods that accurately capture the formation of sharp gradients and cusps in the moving fronts. The algorithms handle topological merging and breaking naturally, work in any number of space dimensions, and do not require that the moving surface be written as a function. The methods can be also used for more general Hamilton-Jacobi-type problems. We demonstrate our algorithms by computing the solution to a variety of surface motion problems.
In this paper, we propose a new model for active contours to detect objects in a given image, based on techniques of curve evolution, Mumford--Shah functional for segmentation and level sets. Our model can detect objects whose boundaries are not necessarily defined by gradient. We minimize an energy which can be seen as a particular case of the minimal partition problem. In the level set formulation, the problem becomes a "mean-curvature flow"-like evolving the active contour, which will stop on the desired boundary. However, the stopping term does not depend on the gradient of the image, as in the classical active contour models, but is instead related to a particular segmentation of the image. We will give a numerical algorithm using finite differences. Finally, we will present various experimental results and in particular some examples for which the classical snakes methods based on the gradient are not applicable. Also, the initial curve can be anywhere in the image, and interior contours are automatically detected.
A general method, suitable for fast computing machines, for investigating such properties as equations of state for substances consisting of interacting individual molecules is described. The method consists of a modified Monte Carlo integration over configuration space. Results for the two-dimensional rigid-sphere system have been obtained on the Los Alamos MANIAC and are presented here. These results are compared to the free volume equation of state and to a four-term virial coefficient expansion.
A general method, suitable for fast computing machines, for investigating
such properties as equations of state for substances consisting of
interacting individual molecules is described. The method consists
of a modified Monte Carlo integration over configuration space. Results
for the two-dimensional rigid-sphere system have been obtained on
the Los Alamos MANIAC and are presented here. These results are compared
to the free volume equation of state and to a four-term virial coefficient
expansion. The Journal of Chemical Physics is copyrighted by The
American Institute of Physics.
In this new edition of the successful book Level Set Methods, Professor
Sethian incorporates the most recent advances in Fast Marching Methods,
many of which appear here for the first time. Continuing the expository
style of the first edition, this introductory volume presents cutting
edge algorithms in these groundbreaking techniques and provides the
reader with a wealth of application areas for further study. Fresh
applications to computer-aided design and optimal control are explored
and studies of computer vision, fluid mechanics, geometry, and
semiconductor manufacture have been revised and updated. The text
includes over thirty new chapters. It will be an invaluable reference
for researchers and students.
Combining unique cytoprobe, rapid hemacyte fractionation, and novel color image analysis, The White IRIS (TWI) extends automated intelligent microscopy to leukocyte differentiation. TWI provides flow cytometry precision and microscopical resolution to review specimens flagged by hematology analyzers with differential capabilities or to complement other analyzers without these capabilities. The system includes compartments for closed sampling, rapid leukocyte-rich plasma preparation, cytoprobe-induced metachromasia, and collection and color analysis of leukocyte images, and presents the results as a single-view 500-cell differential on a 20-in. (50-cm) touch-screen monitor. Method correlations for the five mature cell types averaging r > 0.90 were obtained with a prototype system. Classification of normal and abnormal specimens showed 95% agreement with a reference method without any undetected significant morphologic abnormality. False-positive and false-negative rates of 7.27% and 3.53%, respectively, exceeded performance of current commercial systems. Case studies demonstrate the ease and speed with which unusual pathologies and leukemias can be observed and interpreted.
A fast marching level set method is presented for monotonically advancing fronts, which leads to an extremely fast scheme for solving the Eikonal equation. Level set methods are numerical techniques for computing the position of propagating fronts. They rely on an initial value partial differential equation for a propagating level set function and use techniques borrowed from hyperbolic conservation laws. Topological changes, corner and cusp development, and accurate determination of geometric properties such as curvature and normal direction are naturally obtained in this setting. This paper describes a particular case of such methods for interfaces whose speed depends only on local position. The technique works by coupling work on entropy conditions for interface motion, the theory of viscosity solutions for Hamilton-Jacobi equations, and fast adaptive narrow band level set methods. The technique is applicable to a variety of problems, including shape-from-shading problems, lithographic development calculations in microchip manufacturing, and arrival time problems in control theory.
The morphological appearance of blood cells has an established association to clinical conditions. A novel system, DiffMaster Octavia for differential counting of blood leukocytes, has been evaluated. The system consisted of a microscope, 3-chip color charge coupled device (CCD) camera, automated motorized stage holder, electronic hardware for motor and light control and software for automatic cell location and image processing for preclassification of blood cells using artificial neural networks. The DiffMaster test method, was evaluated on 322 routine blood samples (400 cells per sample) using manual microscopy as reference method. The results showed a correlation of determination (r(2)) of 0.8-0.98 for the normal cell classes and blast cells. The DiffMaster correctly preclassified 89% of all leukocytes with a good reproducibility. After verification of the cell classes, the agreement between the test and reference method was 91% whether the sample was abnormal or normal. The clinical sensitivity was 98% and specificity 82%. The sensitivity to identify blast cells was slightly higher with the DiffMaster than manual microscopy. Similar levels of short-term imprecision for the two methods were found for all cell classes. In conclusion this study shows that the DiffMaster can provide a decision support system which, together with a qualified morphologist, can generate leukocyte differential count reports of high quality.
We propose a new model for active contours to detect objects in a given image, based on techniques of curve evolution, Mumford-Shah (1989) functional for segmentation and level sets. Our model can detect objects whose boundaries are not necessarily defined by the gradient. We minimize an energy which can be seen as a particular case of the minimal partition problem. In the level set formulation, the problem becomes a "mean-curvature flow"-like evolving the active contour, which will stop on the desired boundary. However, the stopping term does not depend on the gradient of the image, as in the classical active contour models, but is instead related to a particular segmentation of the image. We give a numerical algorithm using finite differences. Finally, we present various experimental results and in particular some examples for which the classical snakes methods based on the gradient are not applicable. Also, the initial curve can be anywhere in the image, and interior contours are automatically detected.
Presents a fast segmentation scheme for automatic differential
counting of white blood cells. The segmentation procedure consists of
three phases. First a novel simple algorithm is proposed for
localization of white blood cells. The algorithm is based on a priori
information about blood smear images. In the second phase the different
cell components are separated with automatic thresholding. The
thresholds are selected with a simple recursive method derived from
maximizing the interclass variance between dark, gray and bright regions
based on the method proposed by Otsu (1979). Finally the segmented
regions are smoothed by morphological operations. The segmentation
scheme works successfully for classification of white blood cells. Some
experimental results are also presented
A fast and flexible algorithm for computing watersheds in digital gray-scale images is introduced. A review of watersheds and related motion is first presented, and the major methods to determine watersheds are discussed. The algorithm is based on an immersion process analogy, in which the flooding of the water in the picture is efficiently simulated using of queue of pixel. It is described in detail provided in a pseudo C language. The accuracy of this algorithm is proven to be superior to that of the existing implementations, and it is shown that its adaptation to any kind of digital grid and its generalization to n -dimensional images (and even to graphs) are straightforward. The algorithm is reported to be faster than any other watershed algorithm. Applications of this algorithm with regard to picture segmentation are presented for magnetic resonance (MR) imagery and for digital elevation models. An example of 3-D watershed is also provided
The uniform-cost algorithm is a special case of the A*-algorithm for finding the shortest paths in graphs. In the uniform-cost algorithm, nodes are expanded in order of increasing cost. An efficient version of this algorithm is developed for integer cost values. Nodes are sorted by storing them at predefined places (bucket sort), keeping the overhead low. The algorithm is applied to general distance transformation. A constrained distance transform is an operation which calculates at each pixel of an image the distance to the nearest pixel of a reference set, distance being defined as minimum path length. The uniform-cost algorithm, in the constrained case, proves to be the best solution for distance transformation. It is fast, the processing time is independent of the complexity of the image, and memory requirements are moderate
The results of an automated classification of the peripheral blood leukocytes into eight categories are presented. The classification was achieved by means of digital image processing. The categories were: small lymphocytes, medium lymphocytes, large lymphocytes, band neutrophils, segmented neutrophils, eosinophils, basophils, and monocytes. An eight-dimensional multivariate Gaussian classifier was used. The features were extracted from a 50 Ã 50 point digital image. These features were measures of such visual concepts as nuclear size, nuclear shape, nuclear and cytoplasmic texture, cytoplasm color, and cytoplasm colored texture. The data set consisted of 1041 blood cell images and were divided into a training set of 523 cells and an independent testing set of 518 cells. These cells were digitized directly from the blood smear, which was stained with Wright's stain. Twenty different blood smears were used and were collected over a three-year period from 20 people. The "true" classification of the data set was obtained from four experienced hematology technicians. Their performance was compared to the automated classifier both in terms of an absolute classification of cells and in terms of estimating the percentage composition of the population (or the blood cell differential count). The measure of performance used was the percentage error for each class. The mean percentage error for the eight classes in terms of an absolute classification was 8 and 29 percent for the human observers and the automated classiffier, respectively.
A differential white blood cell classifier is described. The system, which is based upon a three-color flying spot-scanner approach, utilizes recognition parameters based on the principle of geometrical probability functions which are generated at high speed in a dedicated computer. The result is a pattern recognition system capable of performing a routine clinical differential cell count at the rate of 100 cells/40s on a traditionally prepared specimen. The Hematrak system was evaluated in a routine clinical laboratory which reported that reproducibility, normal cell recognition accuracy, and the detection of abnormality were at least comparable to the manual technique.
The simulation of a system to perform an automated leukocyte differential has been achieved through the measurement of certain features or parameters on the cell image. Using photomicrographs on Kodachrome II-A film of Wright's stained leukocytes, four features have successfully separated five normal cell types. Using a sequential procedure based on erythrocyte color to locate the leukocyte in the field of view, statistical analysis on a set of 74 training cells yields a partitioning of the four-dimensional feature space into five distinct regions. An unknown cell may then be classified by measuring its distance to each of the five regions (clusters) via a specified metric.
We present serial and parallel algorithms for solving a system of
equations that arises from the discretization of the Hamilton-Jacobi
equation associated to a trajectory optimization problem of the
following type. A vehicle starts at a prespecified point xo
and follows a unit speed trajectory x(t) inside a region in ℛ<sup>m
</sup> until an unspecified time T that the region is exited. A
trajectory minimizing a cost function of the form
∫0T r(x(t))dt+q(x(T)) is sought. The
discretized Hamilton-Jacobi equation corresponding to this problem is
usually solved using iterative methods. Nevertheless, assuming that the
function r is positive, we are able to exploit the problem structure and
develop one-pass algorithms for the discretized problem. The first
algorithm resembles Dijkstra's shortest path algorithm and runs in time
O(n log n), where n is the number of grid points. The second algorithm
uses a somewhat different discretization and borrows some ideas from a
variation of Dial's shortest path algorithm (1969) that we develop here;
it runs in time O(n), which is the best possible, under some fairly mild
assumptions. Finally, we show that the latter algorithm can be
efficiently parallelized: for two-dimensional problems and with p
processors, its running time becomes O(n/p), provided that
p=O(√n/log n)
An automated classification of blood cells by multistage classifier
- R Suzuki
- A Hashizume
- H Matsushita
- R Yabe
Suzuki R, Hashizume A, Matsushita H, Yabe R. An automated classification of blood cells by multistage classifier. IECON 1984;1152-1157.
Use of watershed in contour detection. Presented at the Proceedings of the International Workshop on Image Processing, Real-Time Edge and Motion Detection/Estimation; Rennes
- S Beucher
- C Lantu
Beucher S, Lantu ejoul C. Use of watershed in contour detection. Presented at the Proceedings of the International Workshop on Image Processing, Real-Time Edge and Motion Detection/Estimation; Rennes; 1979.
A fast segmentation scheme for white blood cell images. 11th IAPR International Conference on Pattern Recognition Conference C
- Cseke I
Segmentation of dense leukocyte clusters In: Proceedings of the Workshop on Mathematical Methods in Biomedi-cal Image Analysis; Kauai, Hawaii
- B Nilsson
- Heyden
Nilsson B, Heyden A. Segmentation of dense leukocyte clusters. In: Proceedings of the Workshop on Mathematical Methods in Biomedi-cal Image Analysis; Kauai, Hawaii; Los Alamtitos, CA: IEEE Computer Society Publications 2001. p 221–227.
Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis
- Nilsson B
- Heyden A
Actes du Second Symposium Européen d'Analyse Quantitative des Microstructures en Sciences des Materiaux
- Digabel H
- Lantuéjoul C
Use of watershed in contour detection. Presented at the Proceedings of the International Workshop on Image Processing
- Beucher S
- Lantuéjoul C
Staining procedures Philadelphia: Lippincott Williams & Wilkins
- G Clark
The white iris leukocyte differential analyzer for rapid high-precision differentials based on images of cytoprobe-reacted cells
- Hl Kasdan
- Jp Pelmulder
- L Spolter
- Gb Levitt
- Mr Lineir
- Gn Coward
- Si Haiby
- J Lives
- Nc Sun
- Fh Deindoerfer
A fast segmentation scheme for white blood cell images
- I Cseke
Cseke I. A fast segmentation scheme for white blood cell images.
11th IAPR International Conference on Pattern Recognition Conference C. Image Speech Signal Anal 1992;3:530-533.
Leukocyte pattern recognition
- I T Young
Young IT. Leukocyte pattern recognition. IEEE Trans Biomed Eng
1972;BME-19:291-298.
Philadelphia: Lippincott Williams & Wilkins
- G Clark
Clark G. Staining procedures. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 1981. p 179.
An automated classification of blood cells by multistage classifier
- Suzuki R