Density imaging using inverse scattering.

Area of Science:

• Ultrasonic tomography
• Inverse scattering methods
• Wave propagation modeling

Background:

• Inverse scattering is a robust ultrasonic tomography technique.
• Most methods neglect density changes, focusing on sound speed and attenuation.
• Literature suggests density can be recovered using inverse scattering.

Purpose of the Study:

• To evaluate two inverse scattering algorithm classes for reconstructing circular cylinders, including density changes.
• To compare the dual frequency distorted Born iterative method (DF-DBIM) and the T-matrix formulation.
• To assess the impact of signal-to-noise ratio and bandwidth on reconstruction accuracy.

Main Methods:

• Investigated two classes of inverse scattering algorithms: separation of density contributions (DF-DBIM) and inversion of the full wave equation (T-matrix).
• Reconstructed circular cylinders with moderate density changes using both DF-DBIM and T-matrix methods.
• Analyzed root mean square error (RMSE) for reconstructions up to eight wavelengths in diameter.

Main Results:

• Both DF-DBIM and T-matrix methods achieved RMSE below 30% for moderate density changes.
• DF-DBIM required high signal-to-noise ratios for accurate reconstructions.
• T-matrix formulation necessitated significantly large bandwidths for accurate results.

Conclusions:

• Inverse scattering methods can reconstruct sound speed, attenuation, and density.
• Practical implementation requires careful consideration of signal-to-noise ratio and bandwidth limitations.
• Both DF-DBIM and T-matrix methods show potential for advanced ultrasonic tomography applications.