Noll (2006) US Notes 1: page 1 Ultrasound Notes, Part 0 – Roadmap • Introduction • Physics: reflections, wave equations, pulse functions, attenuation, dispersion. • The ultrasound signal: bandpass signals, reflections, speckle noise. • 2D imaging: voltage, derivation of the point spread function. • The lateral point spread function and diffraction: steady state, pulsed analyses. • Array systems: sampling, grating patterns • System components • Doppler – if time permits. Ultrasound Notes, Part I – The Basics • Ultrasound imaging is a direct, non-reconstructive form of imaging where image formation is obtained by localizing an ultrasonic wave to a small volume in 3D space. • Two dimension of localization are performed by diffraction (focusing), as in optics. • One dimension is performed by pulsing, as in RADAR. • The ultrasonic wave is produced by electrical excitation of a piezoelectric transducer, which is usually transmitted to the body through an impedance matching gel. • Ultrasonic waves are detected by the same transducer and converted into an electrical signal for processing and display. • A simple system is shown here (from Macovski):Noll (2006) US Notes 1: page 2 - A pulser excites the transducer with a short pulse, which propagates through the body reflecting off of mechanical inhomogeneities. The time relationship between the transmitted pulse and received signal is given by cztecho2= where z is the depth from the transducer and c is the velocity of sound in the body (c ~ 1500 m/s or 1.5 mm/μs) - Reflected signal at time t are associated with mechanical inhomogeneities at depth z = tc/2. - The lateral extent of the pulse echo is determined by the cross-section of the “beam.” This cross-section is depth (z) dependent. - One line of an image can be obtained by recording the reflected signal as a function of time – the “shape” of the response is determined by the pulse properties, transducer properties and diffraction. If we plot this amplitude as a function of time, we get an “A-mode” (for amplitude) scan: - 2D and 3D images can be generated by moving the transducer to difference positions (in x and y), thus moving the “beam.” Images can be generated by making the amplitude of the response a grayscale level (or brightness) and when all scan lines have been acquired presented in the form of an image. This is known as a “B-mode” (for brightness) scan. ANoll (2006) US Notes 1: page 3 simple 2D system (x,z) is shown here: - B-mode imaging can also be done by electronic motion of the beam, as will be discussed in later lectures. - Finally, a single line can be examined over different time frames and also presented as a grayscale image. This is known as “M-mode” (for motion). Source of US signals #1: Surface Reflections In ultrasound, we make images the reflectivity in the body, R(x,y,z). One major source of R is surface reflections, which are generated at the surface of 2 materials.Noll (2006) US Notes 1: page 4 We define the characteristic acoustical impedance as Z = ρc, where ρ is the density of the material (in kg/m3) and c is the propagation velocity of acoustic waves (m/s). Another relationship holds (acoustic Ohm’s law): Z=P/v, where P is the pressure (1 Pa = N/m2 = kg/(m s2)) and v is the particle velocity (m/s). Note the distinction between the microscopic behavior of the constituent particles (v) and the macroscopic behavior of the wavefront (c). For the following analysis, we’ll assume that the incident angle and reflected angles are both zero. (Like a mirror, if the mirror is tilted the reflection will not come back to you – thus in US we typically only see surface reflections from sources that are parallel to the wavefront – this is equivalent to θi= θr=θt=0.) There are a variety of boundary conditions that need to be satisfied. First, the net particle velocity must be the same on either side of the boundary (otherwise, particles will cross the boundary): vi – vr = vt Æ Pi/Z1 – Pr/Z1 = Pt/Z2 Second, the pressures on either side of the boundary must be equal (the boundary will shift to make the pressures match – recall we are working in fluids) Pi + Pr = Pt We can now solve for the reflectivity: 01221122ZZZZZZZPPRir−≈+−== where Z0 is average acoustic impedance in the tissue. Comments:Noll (2006) US Notes 1: page 5 • As we can see R ∈ [1,-1], R<0 indicates that the phase of the reflected signal is reversed and R>0 indicates that the phase of the reflected signal is preserved. • Typically, reflections in the human body (other than tissue/bone or tissue/air interfaces) are very small – on the order of 5% or less (20log(|R|) < -25 dB). • Intensities in US are a relative measure – not based on any absolute or inherent tissue property. Thus, relative differences in Z are what is important in determining reflected intensities. • The intensity of the acoustic wave is I = P2/2Z (energy/time/unit area). Ir/Ii = R2 and It/Ii = (1 – R2) [ A (not very interesting) aside: how does this strange ZPI 2/2= relationship come about. Energy density is: 221221/ vVmvEρ== . (But wait, isn’t the motion sinusoidal so E should be ½ of this? No, the other half the energy is stored as potential energy.) Intensity is the propagation of energy through a surface: energy/(time x area) or cvEcI221ρ== , where c is the propagation velocity of the wave (and associated energy). Using ZPv /= and cZρ=, we get ZPI 2/2=. What about units? Units of pressure: Pa = N/m2 = J/m3 = kg/(m s2); units of impedance = kg/(m2 s) = J s/m4 = etc. I, therefore has units of J/(m2 s). ] Source of US signals #2: Volume scattering On a microscopic level (< λ = c/f ), inhomogeneities in tissue scatter sound waves. We can define a parameter η = volumetric backscatter coefficient = backscatter cross section per unit volume. Backscatter is: - weak (20 dB less than surface reflections) - isotropic (scatters in all directions) - always present (this is an intrinsic property of a tissue) - this will be discussed further (below) in the section on Speckle NoiseNoll (2006) US Notes 1: page 6 Wave Equations Assumptions: - homogeneous medium - non-dissipative medium - longitudinal waves (compression or pressure waves) (fluids will not support transverse or shear waves except at surfaces – the difference is similar to