Automated tool for the detection of cell nuclei in digital microscopic images: Application to retinal images
Jiyun Byun a;b, Mark R. Verardo b;c, Baris Sumengen a;b, Geoffrey P. Lewis b;c,
B. S. Manjunath a;b and Steven K. Fisher b;c;d
a Department of Electrical and Computer Engineering
b Center for Bioimage Informatics
c Neuroscience Research Institute
d Department of Molecular, Cellular and Developmental Biology
University of California, Santa Barbara, CA 93106, USA
B. S. Manjunath a;b and Steven K. Fisher b;c;d
a Department of Electrical and Computer Engineering
b Center for Bioimage Informatics
c Neuroscience Research Institute
d Department of Molecular, Cellular and Developmental Biology
University of California, Santa Barbara, CA 93106, USA
Abstract
Purpose: To develop an automated nuclei detection tool that provides reliable and consistently accurate results for counting cell nuclei.
Methods: We propose a simple yet robust method to analyze large sets of digital micrographs. The nuclei detector design is based on a Laplacian of Gaussian filter. We use the leave-one-out cross validation method for estimating the generalization error, which is then used to choose the model and parameters of the proposed nuclei detector with both fluorescent and dye stained images. We also evaluate the performance of a nuclei detector by comparing the results with manual counts.
Results: When our nuclei detector is applied to previously unseen images of feline retina, it correctly counts nuclei with an average error of 3.67%. Our approach accurately identifies the location of cell bodies. We also test the proposed method with various images and show that it is applicable to a wide range of image types with nuclei varying in size and staining intensity. The nuclei detector is developed as an ImageJ plugin and currently available at UCSB's Center for Bio-image Informatics website (http://www.bioimage.ucsb.edu/software.html).
Conclusions: The proposed method is simple and reliable. It also has widespread applicability to a variety of sample preparation and imaging methods. Our approach will be immediately useful in quantifying the number of cells in large sets of digital micrographs and from high-throughput imaging.
Molecular Vision, vol. 12, pp. 949-960, http://www.molvis.org/molvis/v12/a107/, Aug. 2006.
Node ID: 424 ,
DB ID: 226 ,
Lab: VRL ,
Target: Journal