Computer-assisted image processing for quantitative measurements of fractional bone area A methodologic study

Abstract

Abstract

The purpose of this study was to evaluate a method for computer-assisted image processing to assess quantitatively fractional bone volume in sections of bone biopsies (part I). A second purpose was to apply this method in an experimental study to determine bone loss by measuring fractional bone area in sections of calvarial bone of gastrectomized rats and to compare this with a control group. Stained paraffin sections of skull bone of young rabbits (part I) and rats (part II) were examined. The histologic sections were placed on a microscope. A video camera was connected to the microscope, and the image transferred to a display monitor connected to a PC with dedicated software, performing the measurements. In the first part of the study tests of the reproducibility of the method were performed. The influence of factors such as external illumination and light intensity in the microscope was evaluated by measuring the same area several times at different times (= different illumination) and at different microscope light intensity levels. The automatic method was compared with a manual method, using a Merz grid but automatically calculated. The difference between the means of the manual and automatic measurements was determined with the paired t test. In the second part of the study the unpaired t test was used to determine the differences between the control group and experimental group, both for the manually and the automatically measured values. The data of this study indicate that the automatic image processing method is an efficient tool for determining bone loss by measuring bone area to estimate fractional bone volume in histologic sections. However, the quality of the histologic sections is of paramount importance for precise measurements.