In this paper, we use partial diierential equation based analysis as a methodology for computer-aided cytology. We wish to accurately extract and classify the shapes of nuclei from noisy confocal microscopy images. This is a prerequisite to an accurate quantitative intranuclear (genotypic and phenotypic) and internuclear (tissue structure) analysis of cancerous and pre-cancerous specimens. We study the use of a geometric-driven scheme for improving the results obtained by an automated nuclear segmentation method, followed by object reconstruction and interactive classiication. We build a chain of methods that includes an edge-preserving image smoothing mechanism, an automatic segmentation method, a geometry-driven scheme to reg-ularize the shapes and improve edge delity, and an interactive method to split shape clusters and reclassify them. We test the denoising and shape extraction steps on many real confocal microscope images and measure the improvement when compared to other methods.
[1]
A Liljeborg,et al.
Confocal 3-dimensional DNA image cytometry in thick tissue sections.
,
1996,
Cytometry.
[2]
P. Lions,et al.
Axioms and fundamental equations of image processing
,
1993
.
[3]
D. Pinkel,et al.
Segmentation of confocal microscope images of cell nuclei in thick tissue sections
,
1999,
Journal of microscopy.
[4]
Guillermo Sapiro,et al.
Implementing continuous-scale morphology via curve evolution
,
1993,
Pattern Recognit..