The direct examination of three-dimensional bone architecture in vitro by computed tomography.

We describe a new method for the direct examination of three-dimensional bone structure in vitro based on high-resolution computed tomography (CT). Unlike clinical CT, a three-dimensional reconstruction array is created directly, rather than a series of two-dimensional slices. All structural indices commonly determined from two-dimensional histologic sections can be obtained nondestructively from a large number of slices in each of three orthogonal directions. This permits a comprehensive description of structural variation within a specimen and greatly facilitates the study of structural anisotropy. A measure of three-dimensional connectivity (Euler number/tissue volume) has been determined for the first time in human cancellous bone and shown to correlate with several two-dimensional histomorphometric indices. The method has the potential for overcoming many of the limitations of current approaches to the study of bone architecture at the microscopic level.

Evaluation of a microcomputed tomography system to study trabecular bone structure.

A new microcomputed tomography (micro-CT) system and thresholding procedure was evaluated as a tool for nondestructive analysis of trabecular bone. Images of 6-mm trabecular bone cubes acquired from the micro-CT system were compared with optical images of corresponding histologic sections to determine the accuracy of representation. The stereologic measures of bone volume fraction (PP) and trabecular plate density (PL) were used to quantify the comparisons. The results showed that the micro-CT measures of PP were not significantly different from those measured from histologic sections and therefore were very accurate. Measures of PL were different by approximately 14%, which translated into discrepancies in trabecular plate thicknesses of about 19 microns. This difference was significantly correlated to the microstructural characteristics of the specific specimen scanned. The precision of both measurements was excellent.

Motion of mitochondria in cultured cells quantified by analysis of digitized images.

Translational movements of mitochondria in cultured rat liver cells were characterized quantitatively by using a video camera to detect and a video digitizer-computer system to analyze fluorescent images of mitochondria stained with rhodamine 123. The centroids of the images of individual mitochondria were determined at selected time intervals and the paths followed by the mitochondria were defined by the paths of the centroids. The predominant translation motion of the mitochondria satisfied the formal conditions of a Brownian random walk for a free particle, although in several cases there was a slow drift superimposed on the random motion. The apparent diffusion coefficients were approximately 5 X 10(-12) cm2 s-1, and the drift speeds approximately 2 X 10(-3) micron s-1.