The Relationship Between Ultrasonic Backscatter And The Characteristic Parameters Of Cancellous Bone

Osteoporosis is a musculoskeletal disease mainly characterized by the decrease in bone density and the deterioration of bone microstructure, leading to an increased risk of fracture Osteoporosis is attracting more and more attentions as the aged population continues to grow in modern society. It is estimated that over 200 million people worldwide suffer from osteoporosis. This disease impinges adversely on the quality of life and osteoporotic fractures impose a considerable financial burden on health services. Therefore, the prevention of osteoporosis as well as the early identification and treatment are of great importance. Currently, X-ray-based densitometry techniques, such as quantitative computed tomography (QCT) and dual-energy X-ray absorptiometry (DXA) are used in the clinic to evaluate bone quality and predict the risk of bone fracture. The bone mineral density (BMD) provided by DXA is considered as the gold standard for the clinical diagnosis of osteoporosis. However, BMD alone represents only 60% of fracture prediction. Quantitative ultrasound (QUS) is considered as a good candidate with the advantages of safety, nonionizing radiation, inherently lower cost and portability. QUS for cancellous bone evaluation can mainly be classified into two methods, transmission and backscattering. The backscatter technique uses a single transducer to both transmit and receive ultrasonic signals and can provide easier access to the bones, such as the hip and spine, which are more sensitive to osteoporosis. Backscatter is known to provide information regarding BMD, number density of scatterers (trabeculae), elastic properties and the complex microstructure of cancellous bone. Cancellous bone evaluation based on ultrasonic backscatter technique has become a promising topic in bone ultrasound field.The goals of this thesis are to improve and perfect the cancellous bone ultrasonic backscatter theory, and promote the application of ultrasonic backscatter in cancellous bone evaluation in the clinic. The main contents are summarized as followings:1. Establish a simulation system for ultrasonic backscatter in cancellous boneTwo-dimensional (2D) and three-dimensional (3D) simulation system was established based on the finite-difference time-domain (FDTD) method. The ultrasonic phenomena including scatter, transmission, diffraction and absorption were all taken into account in the simulation system. The perfectly matched layer (PML) was employed to absorb the boundary reflection. The simulation system was a good help in the study of ultrasonic backscatter in cancellous bone.2. Analyze the relationship between ultrasonic backscatter and trabecular material propertiesThe effect of trabecular material properties (density, Lame coefficients, viscosities, and resistance coefficients) on ultrasonic backscatter coefficient (BSC) and integrated backscatter coefficient (1BC) was investigated using the backscatter simulation system. The results show that the BSC is a complex function of trabecular bone density, and the 1BC increases as density increases. The BSC and IBC increase with the Lame coefficients. While not very sensitive to the second viscosity, the BSC and IBC decrease as the first viscosity and resistance coefficients increase. The results demonstrated that, in addition to BMD and microarchitecture, trabecular material properties significantly influence ultrasonic backscatter signals in cancellous bone.3. Analyze the relationship between backscatter and trabecular alignmentCancellous bone is an anisotropic tissue and the trabecular elements usually align in a particular principal direction, referred as main trabecular alignment (MTA). Ultrasonic backscatter measurements were performed on cylindrical bovine cancellous bone samples in vitro. The results showed that the backscatter signal was exquisitely sensitive to the trabeculae alignment, and the signal was the strongest when the ultrasound wave propagated perpendicularly to the MTA. The MTA can be estimated by the maximum of signal energy or the minor axis of fitted ellipse for backscatter parameters versus specimen angle. The eccentricity of the fitted ellipse was highly significantly correlated with the degree of anisotropy (DA) (R=0.92-0.94, p<0.05). The results demonstrated that the MTA orientation and DA of cancellous bone can be estimated by ultrasonic backscatter, providing a new method for cancellous bone evaluation.4. Investigate the selection standard for the ultrasonic backscatter signal of interest (SOI)No specific standard has been proposed for the SOI selection and the investigators usually observed absolutely opposite results for ultrasonic backscatter cancellous bone evaluation (negative or positive correlation). The in vitro experiments demonstrated that:(1) the avoided length (T1) and the selected length (T2) directly affected the apparent integrated backscatter (AIB):AIB decreased with the increase of T1, AIB increased with fluctuations as T2 increased when T2 was short (<2 μs), after then, AIB changed little; (2) SOI directly affected the observed correlation between AIB and BV/TV:positive for relatively short T1 while negative for longer T1 (3～8 μs), and the correlation fluctuated as T2 changed when T2 was below 2 μs. (3) Different correlations between AIB and BV/TV were also observed for the backscatter signals at different central frequencies (0.5～10 MHz):positive correlation was preferred when the frequency was below 2.25 MHz, while negative correlation was preferred at higher frequencies. An explicit standard for SOI selection over large frequency bands (0.5～10 MHz) was proposed for backscatter measurements, which was quite important for ultrasonic backscatter cancellous bone evaluation.5. Investigate the effect of cortical shell on ultrasonic backscatter measurementsThe overlying dense cortex obstructs the ultrasonic evaluation of the underlaid cancellous bone in vivo. The effect of cortical thickness (CTh) on backscatter signals was investigated numerically. At various frequencies (1-3.5 MHz), the AIB decreased with big fluctuations as the CTh increased (R=-0.72～-0.90,p<0.001), and the fluctuations were much bigger when the CTh was below 2.18 mm; after removing the multiple reflections in the cortical shell, the AIB-c significantly decreased (p<0.05) and the fluctuations were slightly smaller; after employing the compensation algorithm, no significant correlations were observed between the CTh and the compensated AIB (CAIB) (R=-0.02～0.01, p>0.05) and the fluctuations in CAIB with CTh were reduced. The results demonstrated that the effect of cortical thickness on backscatter signals can be removed by the proposed compensation method.The studies on the relationships between backscatter and trabecular material properties and trabecular alignment strengthen the theory of the ultrasonic backscatter cancellous bone evaluation. The investigations on the selection standard for backscatter signal of interest and the effect of cortical shell on backscatter solved the critical issues for the in-vivo measurements. The results and conclusions in this thesis greatly furthers the application of ultrasonic backscatter in cancellous bone evaluation and the diagnosis of osteoporosis in the clinic.