Artifacts - What is the Cause
1. Three Major Cause of US Artifacts:
- Nature of Ultrasound Physics: limitations posed by the fundamental laws of ultrasound physics that cannot be changed.
- Operator Error: finally, we, the user, can create more artifacts by using improper settings.
- Image Processing Assumptions: with the current technology, the machine can only generate images by processing the echoes it receives based on a set of pre-defined assumptions that cannot be changed. This is category that we are going to focus on.
2. Fundamental Assumptions of US Imaging
These are the 6 assumptions preset in the US machine. They may seem quite simple and perhaps even counter-intuitive to how we think ultrasound waves really behaves.
These are the 6 assumptions preset in the US machine. They may seem quite simple and perhaps even counter-intuitive to how we think ultrasound waves really behaves.
Assumption 1: Ultrasound Travels in a Straight Line Only
It seems rather trivial, but from this assumption spawns the majority of the problems. The machine calculates the depth of a structure based on the time an ultrasound wave has to travel and return to the transducer.
Ultrasound waves do traverse in a straight path. However, there are many ways to change its course.
It seems rather trivial, but from this assumption spawns the majority of the problems. The machine calculates the depth of a structure based on the time an ultrasound wave has to travel and return to the transducer.
Ultrasound waves do traverse in a straight path. However, there are many ways to change its course.
Unfortunately, there is absolutely no way for the machine to determine whether the pulse and/or echo wave took a discourse from a straight path. In the top diagram, the US beam has changed course, however, from the machine’s perspective, it only sees the beam moving in a straight line. This change in path can affect the displayed location of the structure relative to the true anatomical position.
Assumption 2: Echoes are from a Line-of-Sight Only
The pulse wave strikes the reflector surface of a structure, then the echo will return parallel to the pulse wave. In other words, the ultrasound machine thinks that an echo reflects from ONLY one reflector - it cannot tell if the pulse has bound off more than one reflector, and it cannot tell if an echo has returned from an angle that is NOT parallel to the incident pulse wave.
If this rule is violated, the location of the object as projected onto the monitor will not correlate with its true anatomical location.
The pulse wave strikes the reflector surface of a structure, then the echo will return parallel to the pulse wave. In other words, the ultrasound machine thinks that an echo reflects from ONLY one reflector - it cannot tell if the pulse has bound off more than one reflector, and it cannot tell if an echo has returned from an angle that is NOT parallel to the incident pulse wave.
If this rule is violated, the location of the object as projected onto the monitor will not correlate with its true anatomical location.
Assumption 3: US speed is constant at 1540m/s
There is no way for the US machine to know how fast a pulse and an echo are actually traversing through the different mediums (and what these mediums are). Hence, in order for the processor to calculate distance, an assumption for the speed of US is needed. Most tissue in the body is composed of soft tissue, therefore 1540m/s is chosen.
There is no way for the US machine to know how fast a pulse and an echo are actually traversing through the different mediums (and what these mediums are). Hence, in order for the processor to calculate distance, an assumption for the speed of US is needed. Most tissue in the body is composed of soft tissue, therefore 1540m/s is chosen.
In reality, US velocity is medium specific and fixed when it is traversing through that specific medium. Hence, when the US wave traverses from one medium to another, its velocity will change from v1 to v2.
How does that happen? When a wave transfer from medium 1 to 2, its wavelength changes. Hence, its velocity, also, changes (see equation to the left).
How does that happen? When a wave transfer from medium 1 to 2, its wavelength changes. Hence, its velocity, also, changes (see equation to the left).
Assumption 4: Echo Amplitude is only related to the Reflectivity of the Tissue
From previous Acoustic Basics Module, we know that this simply is not true. Many factors contribute the amplitude of the echo.
For example, when you ultrasound the a structure of interest, the picture is much better defined with improved contrast and brightness when the US beam is perpendicular the structure in comparison to an acute angle. (eg. Needle imaging for procedural guidance: the needle is most hyperechoic when the needle is perpendicular to the US beam. As the beam insonates the needle at an increasing acute angle, the brightness of the needle fades.
From previous Acoustic Basics Module, we know that this simply is not true. Many factors contribute the amplitude of the echo.
For example, when you ultrasound the a structure of interest, the picture is much better defined with improved contrast and brightness when the US beam is perpendicular the structure in comparison to an acute angle. (eg. Needle imaging for procedural guidance: the needle is most hyperechoic when the needle is perpendicular to the US beam. As the beam insonates the needle at an increasing acute angle, the brightness of the needle fades.
Assumption 5: US Beam is 2D
That is to say the US beam profile is flat with only 2 dimensions - most illustrations draw the US beam in accordance to this assumption.
That is to say the US beam profile is flat with only 2 dimensions - most illustrations draw the US beam in accordance to this assumption.
In reality, the transducer produces an US beam that has a Z-axis that renders the beam profile 3 dimensional. This Z-axis is not in the same plane as the XY axis.
Strong reflector that are situated in the Z-axis can result in echoes detectable by the transducer. These signals will, however, be displayed on the XY plane on the monitor as the machine is programmed to assume that the US beam has only 2 dimensions: X and Y.
Strong reflector that are situated in the Z-axis can result in echoes detectable by the transducer. These signals will, however, be displayed on the XY plane on the monitor as the machine is programmed to assume that the US beam has only 2 dimensions: X and Y.
Assumption 6: All Detected Echoes Originate from the Main Beam
The US machine assumes that all echoes originate from the Main Beam. So how can this assumption can be violated?
Consider the illustration: The US images the Heart which is just adjacent to the Focal Zone of the Main Beam profile, therefore it will be displaced on the monitor. Now, the Smiley Face is NOT along the Main Beam's imaging zones, therefore it will NOT be imaged and displayed on the monitor.
The US machine assumes that all echoes originate from the Main Beam. So how can this assumption can be violated?
Consider the illustration: The US images the Heart which is just adjacent to the Focal Zone of the Main Beam profile, therefore it will be displaced on the monitor. Now, the Smiley Face is NOT along the Main Beam's imaging zones, therefore it will NOT be imaged and displayed on the monitor.
In reality, the US transducer emits NUMEROUS beam profiles that complicate imaging. In addition, the transducer emit OFF AXIS
and LOW ENERGY
beams that are referred to as
Side Lobes
and
Grating Lobes.
A strongly reflective structure situated along these off axis beam profiles can reflect a pulse wave and generate a detectable echo. However, given the pre-defined assumption that ALL echoes originate only from the Main Beam, these off axis echoes will be displayed on the monitor AS IF these off axis structures are within the Main Beam's profile.
Note that the Smiley Face is along the Side Lobe's profile and displaced on the monitor - as if it is situated within the Main Beam's profile. However, in reality, the Smiley Face's location is somewhere else.
A strongly reflective structure situated along these off axis beam profiles can reflect a pulse wave and generate a detectable echo. However, given the pre-defined assumption that ALL echoes originate only from the Main Beam, these off axis echoes will be displayed on the monitor AS IF these off axis structures are within the Main Beam's profile.
Note that the Smiley Face is along the Side Lobe's profile and displaced on the monitor - as if it is situated within the Main Beam's profile. However, in reality, the Smiley Face's location is somewhere else.
Take Home Messages:
- Some US Artifacts are iatrogenic - can be attenuated by proper Knobology
- Pre-defined assumptions, a set of rules that models reality for the ease of computation, is programmed into the US machine processor though such simplifications give rise to artifacts