Diagnostic ultrasound

In radiology, ultrasound is used for two major purposes: to make sectional images and to measure blood flow velocities. The ultrasound imaging technique is named ultrasonography (US). The most commonly used ultrasonic flow measurement technique is called Doppler ultrasound, Doppler sonography, or Doppler flow measurement. Ultrasonography (US) is by far the most widespread ultrasound modality in radiology. The use of Doppler ultrasound is steadily increasing, however. The basic principles of ultrasonography are first reviewed.

Ultrasonography

Ultrasound refers to sound waves with a frequency above 20,000 Hz, i.e., above the human hearing range. Frequencies in the 2-10 MHz range are most commonly used (1 MHz = 1 million Hz).

Ultrasonography is performed by transmitting a narrow beam of ultrasound into the body from a transducer. The ultrasound is reflected from the various tissues back to the transducer as echoes. The echoes form the basis of the sectional ultrasound image, quite similar to the sonar of fishing boats. The major steps from transmission of the ultrasound until formation of the final image will briefly be explained.

Transmission and reception of ultrasound

The ultrasound is generated in a hand-held transducer, usually placed on the skin of the patient adjacent to the anatomical region to be examined (Fig. 11). The most essential part of a transducer is one or several piezoelectrical crystals. These crystals have a dual property; application of an electrical potential across the crystal causes mechanical deformation of the crystal, and mechanical compression of the crystal generates an electrical potential. Mechanical vibration of the crystal is induced by a short electrical pulse, and the frequency of the ultrasound thus generated, is determined by the resonance frequency of the crystal, which in turn is determined by the thickness of the crystal. The thinner the crystal, the higher the frequency. The echoes reflected back to the transducer, set up mechanical vibrations in the crystal creating electrical signals of the same frequencies as those of the echoes. In this way the echoes are recorded.

Figure 11. Ultrasound scanner. The handheld transducer is placed on the skin of the patient after application of a gel. The real-time dynamic image is displayed on the monitor. The image may be "frozen" and hard copied, or a dynamic scanning sequence may be recorded on videotape. (Phototechnical Department, Rikshospitalet, Oslo.)

The ultrasound being transmitted from the transducer is pulsed. An ultrasound pulse of 1 s duration is transmitted 1,000 times per second. The remaining 999/1,000 (or 99.9%) of the time, the transducer acts as a receiver, listening for echoes. The creation of echoes is explained in the next section.

Attenuation and reflection

The intensity of the ultrasound transmitted, is gradually reduced by passage through the tissues of the body. The total loss of intensity (or power) is named attenuation. The major reason for the attenuation is absorption of the ultrasound as heat. This part of the attenuation is proportional to the ultrasound frequency; the higher the frequency, the higher the loss of energy as heat.

The part of the ultrasound not being absorbed, may be scattered or reflected from the tissues back to the transducer as echoes. A fraction of
the ultrasound will be reflected whenever there is a change in the "resistance" to the propagation of the ultrasound. How easily ultrasound propagates through a tissue is dependent partly upon the particle mass (which determines the density of the tissue), and partly upon the elastic forces binding the particles together. The elasticity of a tissue largely determines the propagation velocity of ultrasound through that tissue. The density and elasticity of a tissue together determines its so-called acoustic impedance (or "resistance"). (Z = ? · c, where Z is the acoustic impedance, ? is the density, and c is the propagation velocity of the ultrasound in the tissue.)
The larger the change in acoustic impedance, the larger the reflection

of ultrasound. Between soft tissue and gas there is an extremely large difference in acoustic impedance, and nearly all of the ultrasound is reflected at the border. This is why a gel is applied between a patient's skin and the transducer to expel the air that otherwise would have stopped the ultrasound beam, and it is also the reason why ultrasonography is unable to display regions covered by bowel gas or air-filled lung tissue. There is a relatively large difference in acoustic impedance between soft tissue and cortical bone as well, and most bones therefore restrict the application of diagnostic ultrasound.

From echo to image: A-mode, M-mode, and B-mode

Every echo returning to the transducer generates an electrical signal whose strength (amplitude) is determined by the strength of the echo. The transformation of electrical signals into an image on a monitor is based on the relatively constant propagation velocity of ultrasound through tissues. By measuring the time from transmission of the ultrasound pulse to reception of the echoes, the depths from which the echoes originated can be estimated. During the "listening" period after each pulse transmission, numerous echoes are recorded from ever increasing depths. Due to the attenuation of the ultrasound in the tissues, echoes from the deepest structures will be the weakest. This is compensated for by increasing the amplification of the electrical signals generated by late arriving echoes. The later the arrival of the echo (i.e., the deeper its origin), the more amplification applied by the so-called time gain compensation or time gain control (TGC).

The simplest visual display of the recorded echoes is the so-called A-mode (amplitude mode) display. In this format, the echoes from the various depths are shown as vertical spikes on a horizontal line indicating depth (or actually time) (Fig. 12 b). The first recorded echo after a pulse transmission is shown to the far left, and the latest recorded echo is shown to the far right on the depth line. The strength of the echo determines the height or amplitude of each spike shown, hence the term amplitude mode or A-mode. The A-mode format gives only a one-dimensional display of changes in acoustic impedance along the ultrasound beam, and is very seldom used in radiology.

Dynamics may be added to the A-mode format by an alternative method named M-mode (M = motion) or TM (time motion) mode (Fig. 12 c). In this display, the depth axis has a vertical orientation on the monitor. The various echoes are not shown as deflections along a line, but rather as dots having a brightness determined by the echo strength. These bright dots are made to scroll across the screen from left to right, thus creating bright curves indicating the change in position of the reflecting structures with time. The monitor curves are updated each time the bright dots reach the right hand side of the screen (similar to the display shown by ECG monitors). The M-mode curves provide very detailed information on the motional behaviour of reflecting structures along the ultrasound beam, and the method has been especially popular in cardiology to show the motion patterns of the various cardiac valve leaflets.

In radiology, the B-mode (B = brightness) format is used almost exclusively. The term indicates that the echoes are shown as bright dots on the screen, the brightness being determined by the echo strength. A B-mode image gives a two-dimensional, sectional display of the anatomy (Fig. 13).

In the early days of ultrasonography, compound scanners providing static images dominated the market. Today, these have been replaced by real-time scanners. The transducers used in real-time scanning often have numerous small crystal elements arranged side-by-side. By different techniques, a narrow beam of transmitted ultrasound is made to scan or sweep through the patient in a linear (Fig. 13 a) or sector-shaped fashion, and echoes are recorded from each position (scan line) of the beam. One scan line position may correspond to the position of one single crystal element. The echoes from all the scan lines create a rectangular or sector-shaped image on the screen (Fig. 13 b). The image is dynamic and may show phenomena such as respiratory motion, vessel pulsations, cardiac contractions, and foetal movements. The transducer is connected to the ultrasound instrument with a pliable cable to enable any position and angulation of the transducer.

Modern ultrasound scanners employ digital techniques. The analogue electrical signals generated in the transducer crystal by the echoes, are digitised, and a digital image matrix based upon signal strength is created. In the final image shown by the monitor, the pixels are given shades of grey determined by the corresponding numbers in the digital matrix.

Doppler sonography

Measurement of blood flow velocity using ultrasound, is usually based upon the general phenomenon that the frequency of a wave form is dependent upon the relative velocity between the emitter and receptor of the wave. This is the Doppler effect, which is applicable to any kind of wave, both electromagnetic (e.g. light) and mechanical (e.g. ultrasound).

In Doppler sonography of blood vessels, a narrow beam of ultrasound is transmitted into the body from a Doppler transducer. If the ultrasound beam intersects a blood vessel or a cardiac chamber, a small fraction of the ultrasound will be reflected from the red blood cells. If the red blood cells are flowing towards the Doppler transducer, the echoes reflected will have a higher frequency than the one emitted from the transducer. When blood flows away from the transducer, the echoes will have a lower frequency than the emitted one. The difference between the frequency of the echoes received by the transducer and the frequency of the ultrasound emitted from the transducer, is called the Doppler frequency shift, or sometimes just the Doppler shift or Doppler frequency. This shift in frequency is directly proportional to the blood flow velocity. During flow measurement, the Doppler frequency shift is continuously estimated by the Doppler instrument, and most instruments will also automatically convert the change in ultrasound frequency into relative blood flow velocity (e.g. m/s). Relative velocity means the component of the velocity pointing straight towards the transducer. If the angle between the Doppler beam and the direction of blood flow (the so-called Doppler angle) is known, then the true flow velocity may be calculated.

When measuring blood flow velocity, the Doppler frequency shift is usually within the frequency range of human hearing. All Doppler instruments are therefore equipped with loudspeakers making it possible to listen to the Doppler frequency shifts of the blood flow. This "sound of blood flow" is very helpful to the examiner, both in localising vessels and in semi-quantitative assessment of flow patterns and velocity. Of course, this audio display is far too inaccurate for the exact quantification of flow velocity. A visual display of the flow velocity is therefore also provided by the Doppler instrument, usually as a graph or wave form showing velocity along the ordinate and time along the abscissa. In most blood vessels, flow velocity is not uniform across the vessel lumen; most often the velocities are highest in the centre of the vessel and decreasing towards the vessel walls. A so-called full spectral display shows the variation with time of all the flow velocities present in the vessel (Fig. 14, bottom). Single line tracings showing how for example the maximum or mean velocity changes with time, may also be presented.

Principally, there are two ways of transmitting and receiving ultrasound in Doppler applications: continuous wave mode (CW) and pulsed Doppler mode (PD). In the continuous wave made, the Doppler transducer contains two separate crystals. One crystal continuously transmits and the other crystal continuously receives the echoes. This concept allows measurement of very high velocities. Velocities are measured from a large range of depths simultaneously, and it is not possible to selectively measure velocities at a particular predetermined depth. In pulsed Doppler mode, the same crystal both transmits and receives the ultrasound. The ultrasound is transmitted as brief pulses, and the echoes are registered in the waiting time between pulse trans- missions. The time from pulse transmission to echo reception determines the depth at which velocities are being measured. Pulsed Doppler ultrasound makes it possible to measure flow velocities from very small volumes (so-called sample volumes) along the ultrasound beam, but the maximum measurable velocities are considerably lower than those measurable with continuous wave Doppler.

The most commonly used Doppler instruments in radiology are so-called duplex scanners, combining real-time ultrasonography and pulsed Doppler sonography. In duplex scanning, the direction of the Doppler beam is superimposed on the B-mode image, and the size and location of the sample volume along the beam can be selected by means of electronic markers (Fig. 14). When an electronic cursor is manually placed parallel to the direction of blood flow, the Doppler angle is measured automatically, and the true flow velocity is displayed (Fig. 14). If the cross-sectional area of the vessel is measured, volume blood flow may also be calculated (as e.g. ml/s).

Colour flow imaging is a further extension of duplex scanning. Colours are superimposed on the real-time B-mode image to indicate the presence of flowing blood. Stationary tissues are shown in shades of grey, and vessels are given a colour (shades of blue, red, yellow, green) determined by relative mean velocity and direction of flow. The colour coded image gives a good overview of the various vessels and flow directions present, but the quantitative information provided by this method is less accurate than that provided by continuous wave or pulsed Doppler. Colour flow imaging is therefore always combined with pulsed Doppler sonography, and the colour flow image serves as a good guidance for placement of the pulsed Doppler sample volume.

Hans-Jørgen Smith