Comparison of several digital and stereological methods for estimating surface area and volume of cells studied by confocal microscopy

Background: The implementation of different methods for estimating the surf ace area and volume of cells studied by confocal microscopy was developed. The methods were compared from the point of view of their precision, applic ability and efficiency. Methods: Interactive stereological methods (spatial grid method, fakir meth od, Cavalieri principle) as well as automatic digital methods (digital Crof ton method, voxel counting, triangulation method, iso-intensity contouring method) were considered. The methods were tested on model geometrical solid s and on real volume images consisting of a stack of serial sections encomp assing entire tobacco BY-2 cells or cell chains. Results: It is shown that many of the studied methods are very precise when applied to cells of simple or moderately complex shapes. The automatic dig ital methods are fast and precise but their applicability is limited by the necessity to segment automatically the object surface and to find an optim al resolution. This limitation is not present in stereological methods whic h are applied interactively and thus are more time-consuming. Conclusions: The presented implementations of the fakir method and the Cava lieri principle enable interactive, unbiased and efficient estimation of th e cell surface area and volume. The recommended steps for measuring the sur face area and/or volume of objects studied by confocal microscopy are descr ibed. Cytometry 36:85-95, 1999. (C) 1999 Wiley-Liss, Inc.