Fig 1 - uploaded by George Teodoro
Content may be subject to copyright.
Source publication
In this paper, we propose a novel segmentation method for cells in histopathologic images based on a sparse shape prior guided variational level set framework. We automate the cell contour initialization by detecting seeds and deform contours by minimizing a new energy functional that incorporates a shape term involving sparse shape priors, an adap...
Contexts in source publication
Context 1
... Our approach is successfully applied to brain histopathological images for cell segmentation task. Our framework, as shown in Figure 1, consists of a seed detection algorithm for cell contour initialization and an integrated contour deformable model that incorporates region, shape and boundary information. ...
Context 2
... Our approach is successfully applied to brain histopathological images for cell segmentation task. Our framework, as shown in Figure 1, consists of a seed detection algorithm for cell contour initialization and an integrated contour deformable model that incorporates region, shape and boundary information. ...
Citations
... Such information allows better coping with clustered cell nuclei and imaging artifacts (e.g., image noise). Most previous deformable shape models (DSMs) employ non-parametric representations (e.g., [8][9][10]) or explicit parametric representations (e.g., [11][12][13][14]), but convex optimization was not used. ...
... Simple curve fitting does not completely fit the shape of the nucleus. Therefore, Zhang and Yu et al. use the level set and snake method to segment the nucleus [5,6]. This method requires more iterations and calculation time. ...
Automatic screening systems play an increasingly important role in the diagnosis of pathologists. Image measurement and classification are the key techniques of automatic screening systems, which directly determine the performance. The distortion in grey and texture after overlapping nuclei segmentation seriously degrades the DNA content measurement and nuclei classification. In order to solve this problem, this paper presents a new method to reconstruct the pixels in overlapping regions based on the GMM-UBM (Gaussian mixture model–universal background model). In this method, a large amount of data are first used to train a GMM (named UBM). Then, the GMM of each nucleus is derived by maximizing a posteriori adaptation with the UBM and the normal grey value of this nucleus. The grey values are randomly generated by the GMM and filled to the overlapping region, with the offset to fine-tuning the Gaussian components. Finally, the image inpainting algorithm is used to repair the connected region. Experimental results show that this method can effectively recover the nucleus features, such as texture, grey and optical density, and improve the accuracy of nucleus measurement and classification.
... Zhou et al. [12] proposed a UNet++ network with deep supervision nesting and full-scale jump connection, which captures features of different depths for superimposition and integration, achieving the goals of parameter minimization and performance optimization. Zhang et al. [13] proposed a method based on Shape Prior Modeling using Sparse, introducing shape constraints to fully weaken the influence of bright and dark regions on the results and achieve more stable segmentation. Nabil [14] proposed a novel architecture Multiblock to develop MultiResUNet upon modifying U-Net, which solves the problem of multi-resolution analysis and reduce memory usage. ...
Medical image segmentation has the significance of research in digital image processing. It can locate and identify the organ cells, which is essential for clinical analysis, diagnosis, and treatment. Since the high heterogeneity of pathological tissues and the inconspicuous resolution in multimodal magnetic resonance images, we propose a multimodal brain tumor image segmentation method based on ACU-Net network. In the beginning, we preprocess brain images to ensure the balanced number of categories. We adopt deep separable convolutional layers to replace the ordinary architecture in the U-Net to distinguish the spatial correlation and appearance correlation of the mapped convolutional channel. We introduce residual skip connection into the ACU-Net to heighten the propagation capacity of features and quicken the convergence speed of the network, to realize the capture of deep abnormal regions. We use the active contour model to against the image noise and edge cracks, come true the tracking of tumor deformation and solve the problem of edge blur in edema area, so as to divide the tumor core and enhanced necrotic parenchymal area exactly in the abnormal area. In this paper,17926 MRI images of 335 patients in the BraTS 2015, BraTS 2018, and BraTS 2019 datasets are used for training and verifying. Our experiments demonstrate that ACU-Net network has better performance than the other segmentation algorithms in subjective vision and objective indicators when applied to brain tumor image segmentation.
... We create a sparse shape prior library with mode detection by non-parametric clustering over shape manifolds derived from manifold learning. In this way, we can effectively extract representative shape codes from a large number of shape annotations without blindly specifying prior shape number and shape codes in the shape library [59]. To provide scalable pathology image processing, we have developed a highly scalable MapReduce based image analysis framework shown in Fig. 3.5 for whole slide image processing, in which segmentation algorithms can be plugged in. ...
Our project is at the interface of Big Data and HPC – High-Performance Big Data computing and this paper describes a collaboration between 7 collaborating Universities at Arizona State, Indiana (lead), Kansas, Rutgers, Stony Brook, Virginia Tech, and Utah. It addresses the intersection of High-performance and Big Data computing with several different application areas or communities driving the requirements for software systems and algorithms. We describe the base architecture, including the HPC-ABDS, High-Performance Computing enhanced Apache Big Data Stack, and an application use case study identifying key features that determine software and algorithm requirements. We summarize middleware including Harp-DAAL collective communication layer, Twister2 Big Data toolkit, and pilot jobs. Then we present the SPIDAL Scalable Parallel Interoperable Data Analytics Library and our work for it in core machine-learning, image processing and the application communities, Network science, Polar Science, Biomolecular Simulations, Pathology, and Spatial systems. We describe basic algorithms and their integration in end-to-end use cases.
... Gharipour and Liew (2016) used level sets for cell nuclei segmentation and identified individual cell nuclei by analyzing the morphology, but shape and intensity information were used in consecutive steps and not jointly exploited. Zhang et al. (2017) used sparse shape priors in conjunction with object overlap penalty terms. None of the above shape-based methods yield globally optimal solutions. ...
Accurate and efficient segmentation of cell nuclei in fluorescence microscopy images plays a key role in many biological studies. Besides coping with image noise and other imaging artifacts, the separation of touching and partially overlapping cell nuclei is a major challenge. To address this, we introduce a globally optimal model-based approach for cell nuclei segmentation which jointly exploits shape and intensity information. Our approach is based on implicitly parameterized shape models, and we propose single-object and multi-object schemes. In the single-object case, the used shape parameterization leads to convex energies which can be directly minimized without requiring approximation. The multi-object scheme is based on multiple collaborating shapes and has the advantage that prior detection of individual cell nuclei is not needed. This scheme performs joint segmentation and cluster splitting. We describe an energy minimization scheme which converges close to global optima and exploits convex optimization such that our approach does not depend on the initialization nor suffers from local energy minima. The proposed approach is robust and computationally efficient. In contrast, previous shape-based approaches for cell segmentation either are computationally expensive, not globally optimal, or do not jointly exploit shape and intensity information. We successfully applied our approach to fluorescence microscopy images of five different cell types and performed a quantitative comparison with previous methods.
... The final nuclei contours are converged through an iterative contour evolution process. This journal paper extends our earlier work 19,20 through substantial method improvement on shape prior library generation, more comprehensive experiments, more in-depth parameter sensitivity analysis. In detail, these important extensions include: ...
We propose a segmentation method for nuclei in glioblastoma histopathologic images based on a sparse shape prior guided variational level set framework. By spectral clustering and sparse coding, a set of shape priors is exploited to accommodate complicated shape variations. We automate the object contour initialization by a seed detection algorithm and deform contours by minimizing an energy functional that incorporates a shape term in a sparse shape prior representation, an adaptive contour occlusion penalty term, and a boundary term encouraging contours to converge to strong edges. As a result, our approach is able to deal with mutual occlusions and detect contours of multiple intersected nuclei simultaneously. Our method is applied to several whole-slide histopathologic image datasets for nuclei segmentation. The proposed method is compared with other state-of-the-art methods and demonstrates good accuracy for nuclei detection and segmentation, suggesting its promise to support biomedical image-based investigations.
... Region based and machine learning based methods have largely been used in cell segmentation but these methods are not observed to be effective in cluster segmentation. Contour based approaches such as snake models or level set models are the state-of-the-art medical image segmentation methods that are increasingly being used for segmentation in medical microscopic images [12][13][14][15][17][18][19] as well as in other medical imaging applications, say, CT segmentation and brain MRI segmentation [20][21][22][23]. This motivates us to explore level set formulation within the probabilistic framework for plasma cell segmentation including cluster segmentation from microscopic images. ...
Plasma cell segmentation is the first stage of a computer assisted automated diagnostic tool for multiple myeloma (MM). Owing to large variability in biological cell types, a method for one cell type cannot be applied directly on the other cell types. In this paper, we present PCSeg Tool for plasma cell segmentation from microscopic medical images. These images were captured from bone marrow aspirate slides of patients with MM. PCSeg has a robust pipeline consisting of a pre-processing step, the proposed modified multiphase level set method followed by post-processing steps including the watershed and circular Hough transform to segment clusters of cells of interest and to remove unwanted cells. Our modified level set method utilizes prior information about the probability densities of regions of interest (ROIs) in the color spaces and provides a solution to the minimal-partition problem to segment ROIs in one of the level sets of a two-phase level set formulation. PCSeg tool is tested on a number of microscopic images and provides good segmentation results on single cells as well as efficient segmentation of plasma cell clusters.
... We create a sparse shape prior library with mode detection by non-parametric clustering over shape manifolds derived from manifold learning. In this way, we can effectively extract representative shape codes from a large number of shape annotations without blindly specifying prior shape number and shape codes in the shape library [53] . ...
Our project is at Interface Big Data and HPC -- High-Performance Big Data computing and this paper describes a collaboration between 7 collaborating Universities at Arizona State, Indiana (lead), Kansas, Rutgers, Stony Brook, Virginia Tech, and Utah. It addresses the intersection of High-performance and Big Data computing with several different application areas or communities driving the requirements for software systems and algorithms. We describe the base architecture including the HPC-ABDS, High-Performance Computing enhanced Apache Big Data Stack, and an application use case study identifying key features that determine software and algorithm requirements. We summarize middleware including Harp-DAAL collective communication layer, Twister2 Big Data toolkit and pilot jobs. Then we present the SPIDAL Scalable Parallel Interoperable Data Analytics Library and our work for it in core machine-learning, image processing and the application communities, Network science, Polar Science, Biomolecular Simulations, Pathology and Spatial systems. We describe basic algorithms and their integration in end-to-end use cases.
... Such approaches are robust because of the absence of local energy minima. Many cell segmentation methods are based on a continuous, variational framework (e.g., [3,4,5,6,7]), where object contours are represented as level sets of functions. Formulating the evolution of such functions as a convex program assures that a globally optimal solution is found reproducibly for any initialization (e.g., [5]). ...
... To better cope with strong image noise and other distortions, cell segmentation methods were proposed which exploit shape information, like shape-regularized variational level sets (e.g., [4,7]) or statistical shape models (e.g., [6]). Other approaches rely on elliptical models, which are fitted by marked point processes (e.g., [9,2,10]) or snake energy minimization (e.g., [11]). ...
... For NIH3T3, results were previously reported for a convex variational level sets approach [5], which does not use shape information. We performed a comparison with this method and Otsu thresholding, and studied the effectiveness of the location constraint of our method in (7). The results in Tab. 1 show that the location constraint improves the accuracy significantly. ...
Among the human body's organs, the brain is the most delicate and specialized. It is proven that after the heart stops then also brain death occurs within 3 to 5 minutes of death or within 3 to 5 minutes of loss of oxygen supply. A brain tumor is a life-threatening disease that can be detected at any age from an infant to an old person. Though a lot of people did research in the detection and analysis of a tumor, but then also detecting tumors at the early phase is still a much more arduous field in the biomedical study. This paper focuses on the comparative study of various existing algorithms in this field. This paper addresses the challenges and some issues in MRI brain tumor detection which are also addressed in this research.