Quantitative Analysis of Bone Microstructure Using Tomosynthesis
Abstract
The purpose of the project is to determine the potential of Digital Tomosynthesis (DTS) for quantitative analysis of vertebral bone quality and for predicting vertebral bone fragility. The current scope, as a proof of concept study, is to compare parameters derived from DTS to those measured from microCT and test their ability to predict in vitro vertebral strength by a set of destructive tests on vertebrae from human cadavers. As of writing of this report, we characterized the DTS system for its ability to resolve features, shapes and densities using a number of man-made reference materials. We have noted that a single best scan is obtained in a scan direction oblique to the major orientation of the structure of interest but multiple orthogonal scans can provide complementary information. We noted that zero angle view of the scans provides density information complementary to the structural information obtained through the synthesis of all views. By correlating DTS-based parameters of cancellous microstructure (fractal dimension, lacunarity, mean intercept length and line fraction deviation) to those derived from microCT (distribution properties of bone volume fraction, trabecular thickness, number and separation, and degree of anisotropy) and to bone stiffness estimated from large scale finite element analysis, we demonstrated that parameters of cancellous bone microstructure derived from DTS correlate with those derived from microCT and increase prediction accuracy of bone stiffness over bone mass alone. As such, we produced strong evidence that the "concept" being explored, i.e., that quantitative analysis of bone quality using DTS is feasible. Current efforts are focused on i) examination of the information content that is represented in DTS parameters and the nature of the agreement and disagreement between microCT and DTS parameters in order to understand how predictive accuracy of DTS may be increased, and ii) mechanical testing of vertebral bodies.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 2012
- Accession Number
- ADA573941
Entities
People
- Yener N. Yeni
Organizations
- Henry Ford Health