The time of an echo gives the depth of a boundary, d=ct/2. The pulse fades as I=I0e−kx
Lesson 2 of 3Show all lessonsHide
Try the code and the simulations beside the lesson.
From echoes to a picture
In the last lesson we saw how strong one echo is. A scanner turns many echoes into a picture. It needs two facts about each echo: when it comes back and how strong it is.
The A-scan: depth from time
The simplest display is the A-scan, where A stands for amplitude. The probe sends one pulse along one line into the body. The screen draws the strength of the echoes against time.
The pulse travels at speed c to the boundary and back. In the time t it covers twice the depth d:
2d=ct
We want d, so d=2ct.
Two linked pictures. On the left, a probe sends a pulse down through four layers: gel, a layer of fat, a layer of muscle and then bone. On the right, a graph of echo strength against time in microseconds, from 0 to 70. There is a large spike at 0 for the transmitted pulse, a small echo at 19.5 microseconds from the fat to muscle boundary, and a larger echo at 57.4 microseconds from the muscle to bone boundary. A dashed curve shows that the echoes become weaker with time because the pulse is attenuated.
Worked example: reading an A-scan
The picture shows echoes at 19.5μs and 57.4μs. Take c=1540 m/s in both fat and muscle.
First boundary.d1=21540m/s×19.5×10−6s=0.0150 m, so 1.50 cm.
Second boundary.d2=21540m/s×57.4×10−6s=0.0442 m, so 4.42 cm.
The muscle layer. It is 4.42cm−1.50cm=2.92 cm thick.
Check. Each microsecond of echo time is 0.077 cm of depth. The answers are a few centimetres, as they should be. The real muscle speed is 1590 m/s, so the true layer is a little thicker. A scanner assumes one speed, 1540 m/s, for all soft tissue.
The B-scan: a picture made of lines
An A-scan gives only one line. A B-scan (B for brightness) shows each echo as a bright dot. The brighter the dot, the stronger the echo. The probe holds a row of many small crystals, which fire one after another. Each makes one line, and the lines side by side make a picture of a slice of the body. A new picture is made many times each second, so we see a heart valve move.
Attenuation: the pulse fades
As the pulse travels, the tissue absorbs and scatters some of its energy. The intensity falls off exponentially with the distance x:
I=I0e−kx
The attenuation coefficientk, in m−1, depends on the tissue and grows with the frequency. Always count the whole path. An echo from a depth d has travelled 2d, so it is weakened by the factor e−2kd, as well as by the reflection at the boundary.
The trade between detail and depth
Here is the choice a sonographer makes.
A high frequency gives a short wavelength and fine detail. But k is larger, so the pulse cannot reach deep.
A low frequency reaches deep, but the picture is less sharp.
So the eye or a tendon near the skin is scanned at 10 MHz or more, and the liver of an adult at 2 to 5 MHz. A rough rule: soft tissue weakens ultrasound by about 0.5 decibel per centimetre for each megahertz.
To make up for the fading, the scanner amplifies late echoes more than early ones. This is called time gain compensation.
Doppler ultrasound: the speed of blood
Red blood cells reflect a little ultrasound. When the blood moves towards the probe, the echo comes back at a higher frequency. When it moves away, the echo comes back lower. The blood first receives the sound as a moving observer, and then sends it back as a moving source. So the shift is doubled:
Δf=c2fvcosθ
Here f is the frequency sent, v the speed of the blood and θ the angle between the beam and the flow. For blood at 0.5 m/s, a 4.0 MHz beam and θ=0, the shift is about 2.6 kHz. That lies in the range we can hear, so many scanners play it as a sound. A narrowing in an artery makes the blood speed up there, and the shift shows it.
Move the boundary from 2 cm to 8 cm with k=25m−1. By how much does the echo fade? Now double k, as a higher frequency would. What happens at 8 cm?
Practise with new numbers
Try it in Python
The program finds the depth of two boundaries from their echo times, and how much the pulse has faded on its way down and back.
Common mistakes
Using d=ct. The echo time covers the path down and back, so d=ct/2.
Counting the attenuation only one way. The pulse fades on the way back too, over 2d in all.
Leaving out the factor 2 in the Doppler shift. The moving blood shifts the sound twice.
Thinking a higher frequency always gives a better scan. It gives more detail, but less depth.
Remember these
A-scan: echo strength against time. Depth d=ct/2.
B-scan: many lines of bright dots make a picture of a slice.
I=I0e−kx, with x=2d for an echo. k grows with frequency.
High frequency: fine detail, little depth. Low frequency: deep, less detail.
Doppler: Δf=2fvcosθ/c.
Check yourself
Try each one on your own first. If you are stuck, open a hint. Then check with Show the answer.
An echo returns after 91μs in tissue where c=1540 m/s. How deep is the boundary?
A tissue has k=20m−1. What fraction of the intensity is left after the pulse travels 5.0 cm?
Why is the thyroid, close to the skin, scanned at a higher frequency than the liver?
A Doppler probe points along the flow of blood at an angle of 90∘. What shift does it measure?
References
OpenStax College Physics 2e, Chapter 17 Physics of Hearing, section 17.7 Ultrasound, on echoes and Doppler ultrasound.
Cambridge International AS and A Level Physics (9702) syllabus, Medical physics, attenuation of ultrasound.
Enroll free in Ultrasound, Hearing and Sight
Every lesson is open to read. Enrolling also lets you:
Mark lessons done and carry on where you stopped.
Take the tests of this lesson and its course.
Count the course towards the certificate for its chapter.