Echo Signal from Rough Planar Interfaces - Influence of Roughness, Angle, Range and Transducer Type

The received electrical signal from a pulse-echo system insonifying a planar acoustical interface was measured for varying degrees of rms roughness (0 0.16 mm), angle of incidence (typically +/-7 ̊) and range to the transducer. A planar and a focused 5 MHz transducer was used. When insonifying a smooth interface, the normalized spectrum of the received signals for a planar transducer exhibits an increasing number of nulls with increased angle of insonification, as predicted from numerical modeling while the dependence on insonification angle for the focused transducer was smaller and the null pattern was much less distinct. For the planar transducer and for the focused transducer with the interface located at the geometrical point of focus, the energy of the received signal as a function of incident angle was approximately Gaussian with maximum at 0 ̊. For the smooth interface, the 3 dB width was ~0.5 ̊ and ~3 ̊ for the planar and the focused transducer, respectively. 1 . Introduction Tissue characterization has been an active area of research for several decades. The progress has, however, been limited by the fact that the 1D electrical signal from the receiving transducer is the result of not only the 3D distribution of scatterers, the geometry and acoustic properties of interfaces and scatterers, but also to a large extent the field pattern and frequency response of the given transducer and constructive/destructive interference in the backscattered field at the surface of the receiving transducer. As an example, the received electrical signal from a pulse-echo system insonifying a planar acoustical interface depends mainly on surface characteristics, orientation of the interface and on transducer geometry. When the surface characteristics is limited to smooth, the given measurement configuration can be simulated with computer models even when the interface is non-planar. When a rough planar surface is considered, the modeling becomes stochastic in nature, and computer modeling tools are not yet available to readily predict the received electrical signal. To attain an understanding of the received signal from a pulse-echo system as a function of transducer geometry and the roughness properties of the reflector, received signals were acquired for planar interfaces with varying degrees of rms roughness, angle of incidence, and range to the transducer, as depicted in Figure 1. The information derived from such measurements might have potential applications in NDE and on a larger time horizon characterization of pathological changes in tissue interfaces, e.g., artery walls. This quantitative information can also serve as a reference when evaluating future simulation approaches to this problem. Figure 1 Side view of the basic measurement set-up. A rough planar interface with rms roughness Rq and correlation length corr is rotated the angle θa at a distance z0 from a planar or focused transducer (radius of curvature, R) with radius a. 2 . Reflectors and Ultrasound System Four rough and one smooth reflector were made, using four types of sandpaper with different degrees of roughness as molds, and casting with a two component, liquid urethane casting elastomer (Biresin (type U1402, Sika Chemie GMBH, Stuttgart, Germany)). This elastomer has acoustic properties very close to those of human tissues (speed of sound, c = 1450 m/s; density, ρ = 1.06•10 kg/m @ 20 ̊C). The molding took place in vacuum to ensure that the phantom was an exact inverse copy of the sandpaper surface. It was verified that the reflectors contained no voids or particles, so that a single acoustic interface between water and elastomer was obtained a