Browsing by Author "Koyuncu, Can Fahrettin"

Now showing 1 - 6 of 6

Open Access
Automated cell analysis in microscopy images
(2018-09) Koyuncu, Can Fahrettin
High-throughput microscopy systems have become popular recently, which facilitate to acquire boundless microscopy images without requiring human intervention. However, the analysis of such amount of images using conventional methods is nearly impractical since the analysis can take up to months. Additionally, a considerable amount of observer variability may occur since the analysis completely relies on interpretation of the analysts. As a remedy for that, automated decision support systems, which are objective and rapid, have gained more attention. Since these systems conduct analyses at cellular level, they require a cell segmentation model, results of which directly affect the performance of the entire system. There are several challenges in cell segmentation, each of which should be addressed carefully in order to have an accurate cell segmentation model. One challenge is that cells can be grown in multilayer on the plate, which makes them appear as clusters on the image. Segmentation of these cells requires extra effort since they should be splitted from each other. Another challenge is the imperfections on the image such as inhomogeneities of pixel intensities in a cell and insu cient pixel intensity differences at the border of overlapping cells. Yetanother challenge is the heterogeneity in the morphological characteristics of cells. Depending on cell line types, cells may appear in various outlooks. Developing a generic cell segmentation model, which can handle different cells' outlooks and imperfections, is an open and challenging problem. In order to tackle with these challenges, we deal with the cell segmentation problem in two parts: (1)We focus on finding a new representation for microscopy images, helping us simplify the cell segmentation problem, so that imperfections in cells and inhomogeneities in their visual properties can be alleviated, and cell locations can be emphasized better. (2) We focus on developing a more advanced cell segmentation method, with the motivation that it is almost impossible to obtain a perfect representation in practice. Thus, we work on developing more sophisticated cell segmentation techniques that overcome deficiencies on the representation. To this end, this thesis introduces three new cell segmentation models, two of which introduce a new cell representation technique as well. In our experiments, we tested our algorithms on various microscopy images obtained under the uorescence and phase contrast microscopies and compared them with the previous cell segmentation methods. Our experiments show that the proposed algorithms are more effective in segmenting cells and more robust to the aforementioned challenges.
Open Access
Canlı hücre bölütlemesi için gözeticili öğrenme modeli
(IEEE Computer Society, 2014-04) Koyuncu, Can Fahrettin; Durmaz, İrem; Çetin-Atalay, Rengül; Gündüz-Demir, Çiğdem
Automated cell imaging systems have been proposed for faster and more reliable analysis of biological events at the cellular level. The first step of these systems is usually cell segmentation whose success affects the other system steps. Thus, it is critical to implement robust and efficient segmentation algorithms for the design of successful systems. In the literature, the most commonly used methods for cell segmentation are marker controlled watersheds. These watershed algorithms assume that markers one-to-one correspond to cells and identify their boundaries by growing these markers. Thus, it is very important to correctly define the markers for these algorithms. The markers are usually defined by finding local minima/maxima on intensity or gradient values or by applying morphological operations on the corresponding binary image. In this work, we propose a new marker controlled watershed algorithm for live cell segmentation. The main contributions of this algorithm are twofold. First, different than the approaches in the literature, it implements a new supervised learning model for marker detection. In this model, it has been proposed to extract features for each pixel considering its neighbors' intensities and gradients and to decide whether this pixel is a marker pixel or not by a classifier using these extracted features. Second, it has been proposed to group the neighboring pixels based on the direction information and to extract features according to these groups. The experiments on 1954 cells show that the proposed algorithm leads to higher segmentation results compared to other watersheds. © 2014 IEEE.
Open Access
DeepDistance: a multi-task deep regression model for cell detection in inverted microscopy images
(Elsevier, 2020) Koyuncu, Can Fahrettin; Güneşli, Gözde Nur; Çetin-Atalay, Rengül; Gündüz-Demir, Çigdem
This paper presents a new deep regression model, which we call DeepDistance, for cell detection in images acquired with inverted microscopy. This model considers cell detection as a task of finding most probable locations that suggest cell centers in an image. It represents this main task with a regression task of learning an inner distance metric. However, different than the previously reported regression based methods, the DeepDistance model proposes to approach its learning as a multi-task regression problem where multiple tasks are learned by using shared feature representations. To this end, it defines a secondary metric, normalized outer distance, to represent a different aspect of the problem and proposes to define its learning as complementary to the main cell detection task. In order to learn these two complementary tasks more effectively, the DeepDistance model designs a fully convolutional network (FCN) with a shared encoder path and end-to-end trains this FCN to concurrently learn the tasks in parallel. For further performance improvement on the main task, this paper also presents an extended version of the DeepDistance model that includes an auxiliary classification task and learns it in parallel to the two regression tasks by also sharing feature representations with them. DeepDistance uses the inner distances estimated by these FCNs in a detection algorithm to locate individual cells in a given image. In addition to this detection algorithm, this paper also suggests a cell segmentation algorithm that employs the estimated maps to find cell boundaries. Our experiments on three different human cell lines reveal that the proposed multi-task learning models, the DeepDistance model and its extended version, successfully identify the locations of cell as well as delineate their boundaries, even for the cell line that was not used in training, and improve the results of its counterparts.
Open Access
Iterative H-minima-based marker-controlled watershed for cell nucleus segmentation
(John Wiley & Sons, Inc., 2016) Koyuncu, Can Fahrettin; Akhan, Ece; Ersahin, T.; Cetin Atalay, R.; Gunduz Demir, Çiğdem
Automated microscopy imaging systems facilitate high-throughput screening in molecular cellular biology research. The first step of these systems is cell nucleus segmentation, which has a great impact on the success of the overall system. The marker-controlled watershed is a technique commonly used by the previous studies for nucleus segmentation. These studies define their markers finding regional minima on the intensity/gradient and/or distance transform maps. They typically use the h-minima transform beforehand to suppress noise on these maps. The selection of the h value is critical; unnecessarily small values do not sufficiently suppress the noise, resulting in false and oversegmented markers, and unnecessarily large ones suppress too many pixels, causing missing and undersegmented markers. Because cell nuclei show different characteristics within an image, the same h value may not work to define correct markers for all the nuclei. To address this issue, in this work, we propose a new watershed algorithm that iteratively identifies its markers, considering a set of different h values. In each iteration, the proposed algorithm defines a set of candidates using a particular h value and selects the markers from those candidates provided that they fulfill the size requirement. Working with widefield fluorescence microscopy images, our experiments reveal that the use of multiple h values in our iterative algorithm leads to better segmentation results, compared to its counterparts.
Open Access
Object-oriented segmentation of cell nuclei in fluorescence microscopy images
(Wiley-Liss, 2018) Koyuncu, Can Fahrettin; Cetin Atalay, R.; Gunduz Demir, Cigdem
Cell nucleus segmentation remains an open and challenging problem especially to segment nuclei in cell clumps. Splitting a cell clump would be straightforward if the gradients of boundary pixels in-between the nuclei were always higher than the others. However, imperfections may exist: inhomogeneities of pixel intensities in a nucleus may cause to define spurious boundaries whereas insufficient pixel intensity differences at the border of overlapping nuclei may cause to miss some true boundary pixels. In contrast, these imperfections are typically observed at the pixel-level, causing local changes in pixel values without changing the semantics on a large scale. In response to these issues, this article introduces a new nucleus segmentation method that relies on using gradient information not at the pixel level but at the object level. To this end, it proposes to decompose an image into smaller homogeneous subregions, define edge-objects at four different orientations to encode the gradient information at the object level, and devise a merging algorithm, in which the edge-objects vote for subregion pairs along their orientations and the pairs are iteratively merged if they get sufficient votes from multiple orientations. Our experiments on fluorescence microscopy images reveal that this high-level representation and the design of a merging algorithm using edge-objects (gradients at the object level) improve the segmentation results.
Open Access
Smart markers for watershed-based cell segmentation
(2012) Koyuncu, Can Fahrettin
Automated cell imaging systems facilitate fast and reliable analysis of biological events at the cellular level. In these systems, the first step is usually cell segmentation that greatly affects the success of the subsequent system steps. On the other hand, similar to other image segmentation problems, cell segmentation is an ill-posed problem that typically necessitates the use of domain specific knowledge to obtain successful segmentations even by human subjects. The approaches that can incorporate this knowledge into their segmentation algorithms have a potential to greatly improve the segmentation results. In this study, we propose a new approach for the effective segmentation of live cells from phase-contrast microscopy. This approach introduces a new set of “smart markers” for a marker-controlled watershed algorithm, for which the identification of its markers is critical. The proposed approach relies on using domain specific knowledge, in the form of visual characteristics of the cells, to define the markers. We evaluate our approach on a total of 1954 cells. The experimental results demonstrate that the proposed approach is quite effective in identifying better markers compared to its counterparts. This will in turn be effective in improving the segmentation performance of a marker-controlled watershed algorithm.