TY - JOUR
T1 - Adhesion Segmentation Algorithm and Fluorescence Intensity Extraction for Human Umbilical Vein Endothelial Cells
JO - Journal of Fiber Bioengineering and Informatics
VL - 3
SP - 483
EP - 492
PY - 2015
DA - 2015/08
SN - 8
DO - http://doi.org/10.3993/jfbim00144
UR - https://global-sci.org/intro/article_detail/jfbi/4729.html
KW - Human Umbilical Vein Endothelial Cells
KW - Adhesion Segmentation
KW - Nucleus Localization
KW - Label-controlled Watershed Algorithm
AB - Due to the complexity of adhesion and irregularity of human umbilical vein endothelial cells, we propose
a novel cell image segmentation technique by combining cell image with corresponding nucleus image
based on label-controlled watershed algorithm instead of segmenting the cell image directly. Moreover,
nucleus segmentation and location in stained cells are introduced. The most fascinating advantage of the
combined method is its accurate positioning, which ensures that the center of each cell could be positioned
accurately and is conducive to the adhesion and irregular cells segmentation. To determine the activation
time of nuclear factor kappa B and its signaling pathway blocking mechanism, relevant fluorescence
parameters extraction approach is also discussed. Performance and result analysis demonstrate that this
algorithm is valid, precise, and very helpful to further cell analysis and medical diagnosis.