Filtering, reconstruction, and measurement of the geometry of nuclei from hippocampal neurons based on confocal microscopy data.

The cell nucleus is often considered a spherical structure. However, the visualization of proteins associated with the nuclear envelope in rat hippocampal neurons indicates that the geometry of nuclei is far more complex. The shape of cell nuclei is likely to influence the nucleo-cytoplasmic exchange of macromolecules and ions, in particular calcium, a key regulator of neuronal gene expression. We developed a tool to retrieve the 3-D view of cell nuclei from laser scanning confocal microscopy data. By applying an inertia-based filter, based on a special structure detection mechanism, the signal-to-noise ratio of the image is enhanced, the signal is smoothed, gaps in the membrane are closed, while at the same time the geometric properties, such as diameters of the membrane, are preserved. After segmentation of the image data, the microscopy data are sufficiently processed to extract surface information of the membrane by creating an isosurface with a marching tetrahedra algorithm combined with a modified Dijkstra graph-search algorithm. All methods are tested on artificial data, as well as on real data, which are recorded with a laser scanning confocal microscope. Significant advantages of the inertia-based filter can be observed when comparing it to other state of the art nonlinear diffusion filters. An additional program is written to calculate surface and volume of cell nuclei. These results represent the first step toward establishing a geometry-based model of the-dynamics of cytoplasmic and nuclear calcium.