3-DIMENSIONAL METHODS FOR QUANTIFICATION OF CANCELLOUS BONE ARCHITECTURE

Recent development in three-dimensional (3-D) imaging of cancellous bo ne has made possible true 3-D quantification of trabecular architectur e, This provides a significant improvement of the tools available for studying and understanding the mechanical functions of cancellous bone , This article reviews the different techniques for 3-D imaging, which include serial sectioning, X-ray tomographic methods, and NMR scannin g, Basic architectural features of cancellous bone are discussed, and it is argued that connectivity and architectural anisotropy (fabric) a re of special interest in mechanics-architecture relations, A full cha racterization of elastic mechanical properties is, with traditional me chanical testing, virtually impossible, but 3-D reconstruction in comb ination with newly developed methods for large-scale finite element an alysis allow calculations of all elastic properties at the cancellous bone continuum level, Connectivity has traditionally been approached b y various 2-D methods, but none of these methods have any known relati on to 3-D connectivity, A topological approach allows unbiased quantif ication of connectivity, and this further allows expressions of the me an size of individual trabeculae, which has previously also been appro ached by a number of uncertain 2-D methods. Anisotropy may be quantifi ed by fundamentally different methods, The well-known mean intercept l ength method is an interface-based method, whereas the volume orientat ion method is representative of volume based methods. Recent studies i ndicate that volume-based methods are at least as good as interface-ba sed methods in predicting mechanical anisotropy, Any other architectur al property may be quantified from 3-D reconstructions of cancellous b one specimens as long as an explicit definition of the property can be given, This challenges intuitive and vaguely defined architectural pr operties and forces bone scientists toward 3-D thinking. (C) 1997 by E lsevier Science Inc.