Development of a quantitative ultrasound method for the assessment of cortical bone properties using low frequencies ultrasonic guided waves

Abstract

Osteoporosis is associated with a reduction of the bone mass and microarchitectural bone deterioration, leading to a reduction in bone quality and to an increase in fracture risk. The radius fracture is one of the most frequent osteoporotic fractures and can be considered an important site for the early detection of osteoporosis and to prevent future fractures, such as hip and spine fractures. In this context, the axial transmission technique using ultrasonic guided waves was developed to assess bone quality in long bones such as the radius. The method can go beyond a simple estimation of the bone mineral density typically achieved with dual-energy x-ray absorptiometry (DEXA), which is not sufficient to assess fracture risk. Ultrasonic guided waves have the potential to interrogate both the mechanical and geometrical properties of cortical bone. When operating at low frequencies, the sensitivity to variations in intracortical bone properties is improved due to the great penetration depth achieved by the ultrasonic guided wave modes. This is particularly relevant to assess early stages of osteoporosis since early osteoporosis is known to be associated with endosteal resorption. Therefore, low-frequency ultrasonic guided waves hold promises as a rapid, safe and portable screening method for assessment of early osteoporosis in a primary care level. Thus, the purpose of this project is to bring a better understanding of the physical interaction between low-frequency ultrasonic guided waves and the cortical bone structure. In order to do so, a comprehensive and computationally efficient cortical bone model was implemented using the semi-analytical finite element (SAFE) method. The method allows the simulation of ultrasonic guided waves in an irregular, multi-layer and heterogeneous bone cross-section modeled with anisotropic and viscoelastic material properties. The model was applied in the context of axial transmission to investigate the effect of intracortical bone properties on the propagation of low-frequency ultrasonic guided wave modes. The results allowed the identification of a suitable probe configuration as well as a robust signal processing strategy to extract significant mode features from ultrasonic responses. With this information, a prototypical axial transmission configuration was built so that the performance of the method could be evaluated on five ex-vivo radius samples. To do so, a parametrized bone-like SAFE model was implemented into an autonomous model-based optimization routine to perform the inverse determination of different cortical bone properties. The proposed low-frequency axial transmission configuration was able to retrieve reliable thickness and density values for radius specimens as well as provided additional geometrical information associated with the cortical shape of the samples. The predicted properties found were associated to a much larger cortical volume when compared to the conventional inversion techniques using higher frequency. The proposed inverse method has the potential to increase the detectability of early stages of osteoporosis as well as improve the assessment of risk of fracture. The results available can be now used to define the parameters and instrumentation of a pilot clinical study on the detection of osteoporosis.

Date	25 Jan 2019
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Pierre Bélanger (Supervisor) & Julio Fernandes (Co-supervisor)