Robust Nucleus/Cell Detection and Segmentation in Digital Pathology and Microscopy Images: A Comprehensive Review

Xing, Fuyong; Yang, Lin

doi:10.1109/rbme.2016.2515127

Cited by 410 publications

(224 citation statements)

References 300 publications

Supporting

Mentioning

220

Contrasting

Unclassified

Order By: Relevance

“…These images need to be segmented with great accuracy to correctly determine all individual objects (nuclei) and their centers of mass. As summarized in a couple of recent reviews (Meijering, 2012) (Xing and Yang, 2016), a number of methods were developed for the segmentation of nuclei images over the last 50 years. One common approach is used when a shape with several concave points appears in the image after thresholding.…”

Section: Experimental Materials and Methodsmentioning

confidence: 99%

Limits of Applicability of the Voronoi Tessellation Determined by Centers of Cell Nuclei to Epithelium Morphology

et al. 2016

View full text Add to dashboard Cite

It is well accepted that cells in the tissue can be regarded as tiles tessellating space. A number of approaches were developed to find an appropriate mathematical description of such cell tiling. A particularly useful approach is the so called Voronoi tessellation, built from centers of mass of the cell nuclei (CMVT), which is commonly used for estimating the morphology of cells in epithelial tissues. However, a study providing a statistically sound analysis of this method's accuracy is not available in the literature. We addressed this issue here by comparing a number of morphological measures of the cells, including area, perimeter, and elongation obtained from such a tessellation with identical measures extracted from direct imaging acquired by staining the cell membranes. After analyzing the shapes of 15,000 MDCK II epithelial cells under several conditions, we find that CMVT reasonably well reproduces many of the morphological properties of the tissue with an error that is between 10 and 15%. Moreover, cross-correlations between different morphological measures are reproduced qualitatively correctly by this method. However, all of the properties including the cell perimeters, number of neighbors, and anisotropy measures often suffer from systematic or size dependent errors. These discrepancies originate from the polygonal nature of the tessellation which sets the limits of the applicability of CMVT.

show abstract

Section: Experimental Materials and Methodsmentioning

confidence: 99%

Limits of Applicability of the Voronoi Tessellation Determined by Centers of Cell Nuclei to Epithelium Morphology

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Reference [115,116] discussed a wide range of automated image analysis techniques. The state of the arts on automated nulceus/cell detection and segmentation approaches on digital pathology and microscopy are provided by [117]. Recent developments in deep learning [118][119][120], machine learning [121][122][123] and ensemble learning approaches [124][125] have shown good potential to segment the tissue lesions, and analyze the detected tissues.…”

Section: Future Research Challengesmentioning

confidence: 99%

Endomicroscopy for Computer and Robot Assisted Intervention

Zuo

Yang

2017

IEEE Rev. Biomed. Eng.

View full text Add to dashboard Cite

Abstract-Endomicroscopy is a new technique that allows human tissue to be characterized in vivo, and in situ, circumventing the need for conventional biopsy and histology. Despite increased application and growing research interests in this area, the clinical application of endomicroscopy, however, is limited by difficulties in ergonomic control, consistent probe-tissue contact, large area surveillance and retargeting. Recently, advances in high-speed imaging, mosaicing and robotics have aimed to address these difficulties. The development of robot-assisted devices in particular has shown great promises in extending the clinical potential of endomicroscopy. Issues related to miniaturization, adaptation to tissue deformation, control stability, force and position compensation, cost and sterility are being pursued by both research and commercial communities. In this paper, recent clinical and technical developments in different aspects of computer and robotic assisted endomicroscopy interventions including instrumentation, multiscale integration, and high-speed imaging techniques are presented. We further address emerging trends and new research opportunities towards more widespread clinical acceptance of robotically assisted endomicroscopy technologies.

show abstract

“…intensity, edges, etc) in the image I ( x, y ). The position of the snake is represented parametrically by v ( s ) = ( x ( s ) , y ( s )), s ∈ [0 , 1], whose components denote the image coordinates [34], [35]. It is defined to minimize the energy function…”

Section: Muscle Cell Segmentationmentioning

confidence: 99%

AIIMDs: An Integrated Framework of Automatic Idiopathic Inflammatory Myopathy Diagnosis for Muscle

Sapkota

Liu

Xie

et al. 2018

IEEE J. Biomed. Health Inform.

Self Cite

View full text Add to dashboard Cite

Idiopathic Inflammatory Myopathy (IIM) is a common skeletal muscle disease that relates to weakness and inflammation of muscle. Early diagnosis and prognosis of different types of IIMs will guide the effective treatment. Interpretation of digitized images of the cross section muscle biopsy, which is currently done manually, provides the most reliable diagnostic information. With the increasing volume of images, the management and manual interpretation of the digitized muscle images suffer from low efficiency and high interobserver variabilities. In order to address these problems, we propose the first complete framework of automatic IIM diagnosis system (AIIMDs) for the management and interpretation of digitized skeletal muscle histopathology images. The proposed framework consists of several key components: (1) Automatic cell segmentation, perimysium annotation, and nuclei detection; (2) Histogram based feature extraction and quantification; (3) Content-based image retrieval (CBIR) to search and retrieve similar cases in the database for comparative study; and (4) Majority voting based classification to provide decision support for computer aided clinical diagnosis. Experiments show that the proposed diagnosis system provides efficient and robust interpretation of the digitized muscle image and computer aided diagnosis of IIM.

show abstract

Robust Nucleus/Cell Detection and Segmentation in Digital Pathology and Microscopy Images: A Comprehensive Review

Cited by 410 publications

References 300 publications

Limits of Applicability of the Voronoi Tessellation Determined by Centers of Cell Nuclei to Epithelium Morphology

Limits of Applicability of the Voronoi Tessellation Determined by Centers of Cell Nuclei to Epithelium Morphology

Endomicroscopy for Computer and Robot Assisted Intervention

AIIMDs: An Integrated Framework of Automatic Idiopathic Inflammatory Myopathy Diagnosis for Muscle

Contact Info

Product

Resources

About