An automated image analysis method is described for the quantitative characterization of porous biomaterials. The method consists of impregnating porous materials with epoxy embedding medium and polishing planar sections. Contrast was optimized under a scanning electron microscope using backscattered electron images or under a light microscope using reflected light. The images were analyzed using an image analysis system and software developed for this study. The parameters measured on each image feature of the solid and void fractions, frame by frame, were the area fraction, internal perimeter, minimum crossing distance, number of features per area, equivalent circular diameter, perimeter squared per area, and distribution of equivalent circular diameter squared.
The usefulness of this method was validated by comparing the median values of these parameters and their statistical distributions on Replamineform porous biomaterials derived from echinoid and coral skeletons. We believe that this method for quantitative characterization of the three-dimensional structure of biomaterials is applicable to a broad range of materials and is readily amenable to standardization.