Sensitivity of low-frequency axial transmission acoustics to axially and azimuthally varying cortical thickness: A phantom-based study

PLoS One. 2019 Jul 17;14(7):e0219360. doi: 10.1371/journal.pone.0219360. eCollection 2019.

Abstract

Purpose: Cortical thickness (cTh) is one of the main factors determining a bone's mechanical properties, and its quantification is therefore critical for understanding and monitoring bone pathologies such as osteoporosis. Axial quantitative acoustics (ax-QA) offers a non-radiative, non-invasive method to measure cTh. Even though previous works have ascertained ax-QA's ability to measure azimuthally varying cTh, the effect of axially varying cTh remains unclear. Furthermore, previous experiments and theoretical predictions indicate that measurement of the fundamental flexural mode at low frequencies in the kHz range could increase sensitivity to cTh. However, due to the associated long wavelengths, the approximation of bone geometry as a tube could break down at such frequencies. The presented study therefore investigates a) the sensitivity of ax-QA measurements to cTh in the kHz-regime, b) the applicability of tube theory in this regime, and c) the effect of varying cTh along the long axis on the bone.

Materials and methods: Axial-transmission acoustic measurements were performed at 3kHz on 14 bone phantoms with a femur-like cross-section and a) axially varying cortical thickness or b) axially and azimuthally varying cortical thickness (cTh-range: 1.5mm-7.5mm). Experimental results were compared to theoretical predictions based on an exact elastic tube theory.

Results and discussion: Phase velocity measurements using low-frequency ax-QA exhibited a high sensitivity to local cTh less than 4mm, albeit with a complex, not yet understood pattern. Tube theory failed to predict the wave's behavior in the kHz range, indicating that due to the corresponding long wavelengths the bone can no longer be approximated by a tube, thus requiring more faithful modelling of the bone geometry. The fact that results from both types of phantoms were similar (Pearson correlation coefficient: 0.94) further indicates that the slowly varying cTh along the bone's long axis did not strongly affect wave propagation as measured by ax-QA measurements.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Acoustics*
  • Bone Density
  • Femur / anatomy & histology*
  • Femur / physiology*
  • Fourier Analysis
  • Humans
  • Normal Distribution
  • Osteoporosis
  • Phantoms, Imaging*
  • Signal Processing, Computer-Assisted
  • Stress, Mechanical
  • Tibia / anatomy & histology*
  • Tibia / physiology*

Grants and funding

This study was funded by the Robert Mathys Foundation, Bettlach, Switzerland (https://www.rmsfoundation.ch/home.html), and the ordinary research budget extended to WRT by ETH Zurich. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.