Boguslaw Obara - BioImage Informatics

2D confocal projection image of Arabidopsis thaliana adaxial sepal epidermal nuclei.

I'm moving all my projects to Google Code for the Mercurial source repositories :)

Variability in the control of cell division underlies sepal epidermal patterning in Arabidopsis thaliana

Mathematical morphology-based approach for segmentation of cell nuclei has been used. The algorithm work-flow consist the following procedures:

image pre-processing: where the 2D input image is filtered bymorphological closing and opening with disk structuring element of size 3 pixels,
image segmentation: where the filtered image If is thresholded at automatically selected grey level,
image post-processing: where the segmented image It is filtered by morphological closing by reconstruction with disk structuring element of size 3 pixels.

2D confocal projection image of Arabidopsis thaliana adaxial sepal epidermal nuclei

2D confocal projection image of Arabidopsis thaliana adaxial sepal epidermal nuclei.

Nuclei segmentation result

Nuclei segmentation result.

Zoomed region of interest with segmented nuclei boundaries outlined in red

Zoomed region of interest with segmented nuclei boundaries outlined in red.

Collaborators:

Dr Adrienne H. K. Roeder, Center for Bio-Image Informatics, Department of Electrical and Computer Engineering, UCSB, CA, USA
Prof. Elliot M. Meyerowitz, Division of Biology, California Institute of Technology, CA, USA
Prof. B.S. Manjunath, Center for Bio-Image Informatics, Department of Electrical and Computer Engineering, UCSB, CA, USA

Publications:

B. Obara, J. Byun, D. Fedorov, and B. Manjunath, Automatic nuclei detection and dataflow in Bisquik system, in Workshop on Bio-Image Informatics: Biological Imaging, Computer Vision and Data Mining, Santa Barbara, CA, USA, 17-18 January 2008.