A Marker Controlled Watershed Algorithm with Priori Shape Information for Segmentation of Clustered Nuclei
Main Article Content
Abstract
Microscopy cell image analysis is a fundamental tool for biological research. This analysis is used in studies of different aspects of
cell cultures. Visual inspection of individual cells is very time consuming, insufficient to detect or describe delicate changes in cellular
morphology. The main challenges in segmenting nuclei in histometry are due to the fact that the specimen is a 2-D section of a 3-D tissue
sample. The 2-D sectioning can result in partially imaged nuclei, sectioning of nuclei at odd angles, and damage due to the sectioning process.
Furthermore, sections have finite thickness resulting in aggregating or overlapping nuclei in planar images. Hence a set of image objects that
differ considerably from the ideal of round blob-like shapes occur. Their sizes and shapes in images can be irregular. The classic methodology
for cell detection is image segmentation, which is a fundamental and difficult problem in computer vision. Image segmentation is a fundamental
and difficult problem in computer vision. The difficulty in automatic segmentation of images of cells is often uneven due to auto fluorescence
from the tissue and fluorescence from out-of-focus objects. This unevenness makes the separation of foreground and background a non-trivial
task. The intensity variations within the nuclei further complicate the segmentation as the nuclei may be split into more than one object, leading
to over-segmentation. Due to the cell nuclei are often clustered, make it difficult to separate the individual nuclei. Hence an automatic
segmentation of cell nuclei is an essential step in image histometry and cytometry. This paper presents a robust method to segment clustered
overlapping or aggregating nuclei cells using priori information of shape markers and marking function in a watershed-like algorithm. The shape
markers are extracted using adaptive H-minima transform and prior information about the usual shape of normal/pathological nuclei cells. A new
marking function based on outer distance transform, to avoid jagged boundaries of segmented objects is created. Thus the right sets of markers
and a marking function are used to accurately separate clustered nuclei.
Â
Â
Keywords: Active contours, aggregating/overlapping nuclei, adaptive H-minima, markers, marking function, watershed segmentation.
Downloads
Article Details
COPYRIGHT
Submission of a manuscript implies: that the work described has not been published before, that it is not under consideration for publication elsewhere; that if and when the manuscript is accepted for publication, the authors agree to automatic transfer of the copyright to the publisher.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work
- The journal allows the author(s) to retain publishing rights without restrictions.
- The journal allows the author(s) to hold the copyright without restrictions.