Hide/Show Apps

Human tibial bone strength prediction by vibration analysis for diagnosing progressing osteoporosis

Bediz, Bekir
Osteoporosis is a metabolic bone disease that needs to be properly diagnosed. The current diagnosing procedure of osteoporosis is based on the mineral density of bones measured by common methods such as dual energy X-ray absorptiometry (DXA). However, due to the deficiencies and limitations of these common methods, investigations on the utilization of other non-invasive diagnosing methods have been executed. For instance, using vibration measurements seems to be a promising technique in diagnosing metabolic bone diseases such as osteoporosis and also in monitoring fracture healing. Throughout this study, bone structural modal parameters obtained from vibrations experiments with decreasing mineral density are examined and therefore, it is aimed to find a new approach to detect osteoporosis or progressing osteoporosis by investigating a relation between structural dynamic properties and mineral density of bone. The main advantage of this study is that loss factor, which is an inherit property of bone, is investigated since in the previous studies mainly the changes in natural frequency of bones with the state of osteoporosis is examined. In this thesis, both in vitro and in vivo experiments are carried out on human tibia specimens. The measured frequency response functions (FRFs) are analyzed using modal identification techniques to extract the modal parameters of the human tibia. The results obtained from in vitro experiments show that loss factor may be a powerful tool in diagnosing osteoporosis, however due to the difficulties encountered in the case of in vivo experiments makes the use of this parameter as a diagnosing tool difficult. It is also seen from in vivo experiments that there is a weak correlation between the natural frequencies of tibia and BMD measurements of patients. Therefore, in order to investigate the parameters affecting the natural frequencies of tibia, finite element (FE) model of human tibial bone is constructed. Using this FE model tibia, the effect of boundary conditions of experiments and geometry of the bone on natural frequencies of bone is examined. These analyses show that the effect of both boundary conditions and geometry of tibia is very high. Therefore, it is concluded that if the necessary conditions are satisfied, the using natural frequency information of tibia seems to be a possible and practical method that can be used to detect progressing osteoporosis. Also, using the FE model of tibia, the changes of natural frequencies of tibia with the variation in elastic modulus are investigated.