Sinogram enhancement for high-resolution ultrasonic transmission tomography using nonlinear anisotropic coherence diffusion

A novel system for high-resolution ultrasonic transmission tomography (HUTT) has been developed with the critical innovation of using sub-millimetre transducer arrays and multi-band analysis of the first-arrival pulse. To obtain frequency band-specific projection data or sinograms, the extracted first-arrival pulse at each azimuthal and angular location of a mechanical tomographic scanner is processed to yield a "multispectral" vector of attenuation values at multiple frequency bands. However, the HUTT sinogram is subject to artifacts and noise caused by various sources (e.g., electronic and measurement noise, beam hardening, beam starvation, and ultrasonic diffraction and refraction). The sinogram also degrades at higher frequency bands due to greater attenuation, resulting in low signal-to-noise ratio (SNR). In this study, we present novel methods of enhancing and denoising the sinogram through a variant of nonlinear anisotropic coherence-enhancing diffusion. We demonstrate the performance of various diffusion filters and their effectiveness in improving the quality of the sinogram.