2011 IR&D Annual Report

Quantitative Broadband Ultrasound Techniques to Assess Fracture Risk in Bones, 14-R8082

Principal Investigators
Jorge O. Parra
Qingwen Ni
Todd Bredbenner
Adam Cobb
Daniel Nicolella

Inclusive Dates:  07/07/09 – 12/30/10

Background — The state of the art of modeling ultrasound wave propagation in bone cannot explain the intrinsic attenuation caused by wave-induced flow. This is one of the main motivations to implementing more appropriate poroelastic mechanisms to understand the attenuation of broadband ultrasound propagation in bone. A quantitative understanding of ultrasound interaction and attenuation in bone would provide new capabilities for characterizing osteoporosis. Osteoporosis is a widespread public health problem; in the U.S. alone, 20 million individuals already have the disease and an additional 40 million have low bone mass. Currently, there is no cure for this disease. Early diagnosis of osteoporosis with a technique that can monitor bone quality and predict fracture risk is extremely important for both the treatment and prevention of osteoporosis. The proposed effort is based on advanced models of poroelastic mechanisms. These models have been applied to ultrasonic interrogation of saturated clean sandstone, which has similar properties to cortical bone, and has been successful in predicting observed attenuation. To assess the presence of micro-cracks caused by micro-damage, the Biot squirt-flow mechanism will be applied for the first time to evaluate attenuation measurements in either cortical or trabecular bone, and the estimation of bone permeability from attenuation will be a significant advance in the state of the art. An experimental research program of in vitro ultrasonic measurements of bone samples will support developing modeling and related signal-processing techniques. Ultrasonic measurements with both longitudinal and shear modes of propagation will be made and compared to the predicted values. Micro-damage will be induced in the cortical bone and trabecular bone samples to provide a baseline for a correlation between microscopic structural properties and the modeled and measured quantitative ultrasound data.

The purpose of this research project is to develop a method based on broadband ultrasonic measurements, processing and modeling to assess bone quality based on frequency-dependent attenuation. The overall goal will be to relate wave attenuation/dispersion and wave signal characteristics to bone mechanical properties, effective micro-damage, and permeability in trabecular and cortical bone. The long-term goals of this project will be to develop a methodology to assess bone quality and fracture risk in vivo. Successful execution of this project may provide a basis from which existing clinical bone health diagnostic techniques can be improved or new techniques developed.

Approach — SwRI researchers will perform broadband ultrasound experiments using p-wave and s-wave transducers in bone tissue, without and with induced micro-damage. Ten cortical and 10 trabecular bone samples will be selected to extract mechanic properties and wave attenuation. These properties and attributes will be complemented with other physical properties such as density, dynamic permeability and porosity. The result of the measurements (i.e., attenuation, phase velocity and the wavefield) and physical properties will be cataloged, and will be analyzed qualitatively and statistically and with computer models. The goal is to develop a methodology to assess bone quality from wave attenuation data by relating local-flow length to the effective bone porosity, pore size distribution and induced micro-damage. The results will be verified using NMR data.

Accomplishments — SwRI researchers have demonstrated the use of poroelastic modeling to successfully simulate spectral and attenuation signatures of bones and to show how the displacement of water relative to the bone structure can affect the wave spectrum. The amplitude spectrum will be strong in highly permeable and porous bones, and small in low porosity and low permeability bones. As a result, the wave attenuation for normal bones is greater than the wave attenuation for damaged bones. The theoretical results agree with the observed spectral responses, ultrasound signatures and attenuation curves. This suggests that a procedure can be implemented to process ultrasound data to identify damaged bones using this concept. To implement such a procedure will require improving SwRI’s existing ultrasound system.

Computer simulations to calculate attenuation signatures can be improved by having the experimental source (transducer) function response. This will allow a better fit between observed and calculated attenuation curves. Based on the ultrasonic results, further work needs to be performed on the measurement system to improve future results:

  • Redesign the fixture to include simultaneous ultrasonic measurements, both with and without the bone in place (reference waveform). Measurements performed in this fashion would completely eliminate any issues related to temperature.

  • Include both pulse-echo measurements (with the two transducers acting as transmitters and receivers, in addition to the pitch catch configuration used for attenuation). This would allow for additional ultrasonic measurements of bone thickness.

  • Switch to a smaller transducer diameter to reduce the beam width of the ultrasonic waves. This would further protect against diffraction effects.

  • Discontinue examination of the shear wave measurements. While interesting in a research sense, the practicality of measuring attenuations and velocities in vivo is limited, given that the sound must propagate through a large amount of water, which does not support shear wave propagation.

  • Improve the degassing procedure to ensure that trapped air bubbles do not affect measurements.

Figure 1. Amplitude spectra of bone sample 47754-3 compared with the water-only spectral response (black line). The amplitude spectrum of the pre-damage sample is shifted toward the low frequencies, and the amplitude spectrum of the post-damage sample follows the amplitude spectrum of the water. This represents the change in velocity between the normal and the damaged bone
Figure 1. Amplitude spectra of bone sample 47754-3 compared with the water-only spectral response (black line). The amplitude spectrum of the pre-damage sample is shifted toward the low frequencies, and the amplitude spectrum of the post-damage sample follows the amplitude spectrum of the water. This represents the change in velocity between the normal and the damaged bone.

Figure 2. As the bone becomes damaged (i.e., higher porosity and permeability), its spectral signature approaches the water amplitude response (green curve). The trabecular bone samples have permeabilities of several Darcies and porosities more than 80 percent.  Thus, as the bone becomes damaged, the wave-induced fluid pressure increases toward the water pressure. In the case of a normal bone (lower porosity and permeability), the wave-induced fluid pressure decreases.
Figure 2. As the bone becomes damaged (i.e., higher porosity and permeability), its spectral signature approaches the water amplitude response (green curve). The trabecular bone samples have permeabilities of several Darcies and porosities more than 80 percent. Thus, as the bone becomes damaged, the wave-induced fluid pressure increases toward the water pressure. In the case of a normal bone (lower porosity and permeability), the wave-induced fluid pressure decreases.

Benefiting government, industry and the public through innovative science and technology
Southwest Research Institute® (SwRI®), headquartered in San Antonio, Texas, is a multidisciplinary, independent, nonprofit, applied engineering and physical sciences research and development organization with 9 technical divisions.
04/15/14