MEASUREMENT SCIENCE REVIEW, Volume 5, Section 2, 2005 Image Registration in the T 2 * Measurements of the Calcaneus Used to Predict Osteoporotic Fractures 1 I. Strolka, 2,3 R. Toffanin, 4 G. Guglielmi, 1 I. Frollo 1 Institute of Measurement Science, Slovak Academy of Science, Dúbravská cesta 9, Sk-841 04 Bratislava, Slovak Republic 2 PROTOS Research Institute, P.O. Box 972, I-34012 Trieste, Italy 3 ARCHES, Via G. Leuzzi 18, I-70013 Castellana Grotte, Italy 4 IRCCS - "Casa Sollievo della Sofferenza", I-71013 San Giovanni Rotondo, Italy Email: umerstro@savba.sk Abstract The conventional criterion for fracture risk assessment is the bone mineral density (BMD) measured by X-ray absorptiometry. Even if there is a strong association between bone strength and BMD, nowadays it is well accepted that this is not a sufficiently reliable predictor of fracture risk in osteoporotic patients. Therefore, there is a growing need for a better predictor of bone strength. Several studies have confirmed that the T 2 * relaxation time estimates based on magnetic resonance imaging (MRI) show a strong correlation with bone strength. In this paper, an image-processing tool for the reduction of motion artifacts in the T 2 * measurements of the bone marrow of the calcaneus is described. Keywords: osteoporosis, magnetic resonance imaging, image registration 1. Introduction The importance of noninvasive methods for fracture risk assessment is emphasized by the dramatic increase of osteoporosis incidence in elderly population. Osteoporosis is a metabolic bone disease leading to bone mass reduction and deterioration of the trabecular bone architecture with a consequent increase in bone fragility and susceptibility to fracture. There are two different types of bone present in the skeleton: cortical bone and trabecular bone. The dense cortical (compact) bone creates the rigid shell of bones while the trabecular (spongy) bone is found at the end of long bones, in vertebral bodies and flat bones. The trabecular bone constitutes a complicated three-dimensional structure of interconnected trabeculae (plate- and rod-like elements of the structure having a thickness approximately 0.1-0.2 mm), which is optimized for strength and lightness. The strength of this structure is determined not only by the amount of the bone mineral (which is assessed by the BMD parameter) but morphometric and topologic properties of the structure affect the bone mechanical properties to a significant degree as well. In fact, only about 60% of the bone strength variation can be explained by variations in bone mineral density. A clinical consequence of this is a large proportion of patients with normal BMD scores suffering osteoporotic fractures. Therefore, there is a need for a better bone strength predictor, which takes into account the bone structure arrangement as well. Several studies [2, 3] have demonstrated that the effective transverse relaxation time T 2 * determined by magnetic resonance imaging (MRI) is significantly increased in the osteoporotic bone thus showing the potential of this technique in fracture risk assessment. A recent study at 3.0 Tesla [4] has confirmed that the calcaneus is a suitable surrogate site to assess vertebral osteoporosis and that T 2 * is sensitive to alterations in bone quality not captured by density. The magnetic resonance signal of the mineralized bone decays too fast to be measured using common MRI protocols. However, the trabecular bone cavities are occupied by bone marrow, having relatively long relaxation times T 1 and T 2 . The bone marrow consists mainly 78