Beam Quality describes the shape of the energy spectrum (i.e. the energy distribution of the x-rays) and beam quantity describes the total intensity of the spectrum (i.e. the area under the x-ray spectrum curve). In this post we describe the factors affecting beam quality including: kVp, target material, and pre-patient collimation. We also discuss the mA (tube current) which is the most important factor that affects only the beam quantity and not beam quality.
What is beam quality vs beam quantity?
Before we dive too deep into the details about beam quality and how to measure it we first want to define at a high level, what is beam quality?
Beam quality describes the shape of the x-ray spectrum. So we will review x-ray spectrum briefly here. The x-rays coming out of our clinical x-ray tubes are not all one energy. That would be called monoenergetic or monochromatic. We have a separate post where we describe the physical mechanisms which responsible for x-ray generation. For more details please see that post if you aren’t familiar with the general shape of the x-ray spectrum.
At a high level the x-ray spectrum is a plot of the Number of X-ray Photons (y-axis) as a function of the energy level in keV (x-axis). As we describe in the x-ray generation post the highest energy in an x-ray spectrum is determined by the kVp. The lowest energy photons are typically filtered out by the internal filtration. The peaks in the spectrum are due to characteristic radiation and only contribute a small amount to the total number of photons.
In the remaining sections of the post we will discuss: why we are interesting in beam quality, what affects beam quality and finally how to measure beam quality. In the section on factors that affect beam quality we will spend some time on the difference between beam quality and beam quantity.
We also want to differentiate beam quantity from beam quality where changes in beam quantity may occur without changes in beam quality. Changes in beam quantity mean that the shape of the curve remains the same but there are more or less x-ray photons. The most common way that the beam quantity is changed is via changing the mA (tube current)
- Beam Quality – describes the shape of the spectrum
- Beam Quantity- describes the number of x-rays and beam quantity can be changed without changing the shape, for instance by changing the mA
The area under the plot of the x-ray spectrum is essentially showing how many x-rays are in the beam. If you look at one given energy, i.e. go to a point on the x-axis that corresponds to the energy of interest and then draw a line upwards the point where it intersects with the plot will tell you how many photons this spectrum has at that energy.
The terminology which is used to describe the shape of the x-ray spectrum is soft vs hard:
- A soft x-ray spectrum has more lower energy photons
- A hard x-ray spectrum has more higher energy photons
In this figure there are a couple example x-ray spectrum which correspond to a soft beam and a hard beam, where the harder beam contains a higher proportion of high energy photons.
So, it’s as simple as that. As your beam gets harder, it has a higher proportion of high energy photons. X-ray beams typically get harder when the lower energy photons are attenuated.
Rad take home points:
Beam quality is based on the shape of the x-ray spectrum.
Harder x-ray beams have a higher proportion of more higher energy photons than soft beams.
As professionals in the radiography field (e.g. radiologic technologies or radiographers), you really care about beam quality as it is one of the primary factors that effects the image quality in radiography and CT imaging.
The beam quality will inherently effect the contrast in an image. In general larger image contrast occurs when there is more of a difference between the attenuation coefficients of the different types of material in the image. A softer x-ray beam has a lower average energy and will give you higher image contrast (especially for bone, barium or iodine imaging since these materials have significantly higher attenuation coefficients at low energy).
Despite, this major advantage of soft x-ray beams there are many situations where a harder x-ray will be needed for your clinical imaging scenario. The most significant advantage of the harder beam is the increase in x-ray penetration (i.e. the ability for the x-rays to all pass through the object.) Having sufficient penetration is required to have an adequate exposure on the detector plane. If there is not sufficient penetration the image will be noisier and the clinical diagnosis may be impacted.
Harder x-ray beams also have benefit in terms of artifact reduction, such as metal artifacts or photon starvation artifacts. A harder x-ray beam will also typically have less beam hardening artifacts.
Rad take home points:
- Softer x-ray spectrum has better contrast
- Harder x-ray spectrum has better penetration through thick anatomy
What factors affect beam quality?
Now that we have covered what beam quality is and why we care about it we will cover the factors that effect beam quality. We list them here in a table form for your easy reference. If you are studying you can make flashcards or cover up part of the screen to test yourself.
Easy to change
Higher kVp, Harder Beam
Electrons have more energy, make X-ray photons with more energy on average
X-ray Target Material
Higher Z, Harder Beam
X-ray photons more likely to be attenuated leaving the target
More Filtration, Harder Beam
Lower energy x-rays are more likely to be attenuated by filter, leaving higher energy photons.
In addition to the table we go through briefly with some additional pictures to demonstrate the relationship of different technical factors with beam quality.
kVp Influence on Beam Quality
The first and most obvious technical factor that influences beam quality is the kVp (kiloVolt potential). This is a parameter which is easy to change on the system and a major control knob for the image quality in x-ray radiography and CT imaging.
The reason that the kVp has such a significant impact on the beam quality (i.e. the x-ray spectral shape), is that if you’re accelerating from the cathode to the anode with a higher potential the electrons will have a higher energy when they collide with the anode (target) material. Given a higher incident electron energy the Brememstralung interaction will enable higher energy x-rays to be generated.
This figure demonstrates a schematic of two x-ray spectrum one at a lower energy of 80 kVp and one at a higher energy of 140 kVp. In each case the highest possible energy x-ray that can be generated is determined by the kVp (so 80keV for 80kVp and 140 keV for 140kVp). We have all shaded the region under the curve to demonstrate that if everything else is the same that there are many more x-rays generated by the higher energy spectrum. Since there are more x-rays and the average energy of the higher kVp spectrum is higher this spectrum will penetrate thick anatomy better than the low kVp spectrum.
Target material Influence on Beam Quality
The target material, or the material that the x-ray anode is made of is not a technical parameter that you will be changing on the system itself. There’s not a button typically that says target material and you change the actual target material of the anode. This would be a super complicated design as the anode is the rotating portion of the x-ray tube so it is much harder to change than the cathode for instance.
Pre-patient Filtration Influence on Beam Quality
Another technical factor which has significant impact on the beam quality is the pre-patient filtration. On some radiography systems you will have some control over the pre-patient filtration. On CT scanners the pre-patient filtration is typically designed to even out the x-ray exposure on the detector. Therefore, there will be less filtration in the middle where the patient is usually the thickest and more filtration at the outside where there is less attenuation from the patient. The top view of a profile that is more narrow in the middle and wider on the outside looks like a bowtie. Therefore, standard pre-patient attenuators in CT are called bowtie filters.
After passing through a pre-patient filter the x-ray beam is hardened because the lower energy photons are preferentially attenuated by (i.e. stopped within) the pre-patient filter. For instance, on many fluoroscopy systems a sheet of copper (Cu) is used with thickness of less than 1mm.
Beam filtration is demonstrated in this figure, if we look at an input beam. Imagine we have a spectra that looks like the blue curve. That’s what our input spectra looks like before the impact of the beam filter. That beam filter could be copper, tin, aluminum, etc. Then after when the beam comes out we have the green curve. The vertical dotted lines in blue and green indicate the input and output average energy of the x-ray spectrum. You can see that the average energy is increased after passing through the pre-patient filter.
The main purpose of these type of filters is to reduce unnecessary radiation exposure to the patient. Very low energy photons for instance are not going to be able to penetrate the patient and therefore don’t stand a chance of contributing to the image. Thus, very low energy photons are consider wasted dose as they deposit energy in the patient but don’t contribute to an image.
That is the reason why you would want to have some inherent (i.e. built in) filtration in radiography systems. One source of pre-patient attenuation that isn’t always apparent is the window of the x-ray tube as the x-rays are generated inside of a vacuum sealed envelope. As the x-rays pass through the window in the x-ray tube there is some preferential attenuation of low energy photons.
Is beam quality a function of tube current (mA)?
The figure gives it away in a little title there. In the figure we are pointing out that the mA modulate the overall number of x-ray photons in the spectrum, but it does not change the shape of the spectrum. It’s just scaling this curve up and down but not affecting the shape of this curve at all.
So, beam quality actually DOES NOT depend on the mA. But because we’re changing the number of photons, we want to have a way to describe that as well. So, instead of calling that beam quality, we’ll call it beam quantity. The beam quantity is the number of photons that are under your spectrum. So, here’s two spectra. I drew them by hand, not terribly sophisticated drawing so they might not be perfect. But the idea here is that the spectra should look the same and the only difference is that the one with 200mA has more x-ray photons underneath. But the average energy is the same because the shape of the spectra is the same.
Rad take home points:
- Beam quality depends on: kVp, target material and pre-patient beam filtration.
- Beam quantity is affected by mA but beam quality is dependent on mA.
How do we measure beam quality?
Now that we have established what beam quality is, why we care about it and what factors affect it we will discuss how we can measure beam quality. We demonstrated that the beam quality plays a major influence on the image quality in x-ray and CT imaging. Thus, we want to have a means to ensure that the system is delivering an x-ray spectrum which is similar to the spectrum that we expect.
In the clinic if we want to do quality assurance measurements on the system, how can we measure the beam quality?
Can we look at the signal in the detector to determine the x-ray spectrum? Unfortunately, we can not as the detectors we use they typically measure all of the energy under this spectrum. We can’t just use measurements from our detector to get a measurement of beam quality.
We need a metric to measure the beam quality and a direct measurement is not feasible in the clinic. Instead of doing a direct measurement, we look for a surrogate measurement of beam quality. We want a quantity that we can measure that we can measure in the clinic and is directly related to beam quality.
The surrogate metric which is related to the beam quality is called the ‘half value layer’ (HVL). Half value layer is not a direct measure where we put a detector in and we measure the beam quality itself. Rather we are measuring something that’s related to the beam quality but the HVL is just one number and does not describe the x-ray spectrum completely.
To measure the half value layer, we show in this figure that you can use an ion chamber to measure the exposure. We’ve covered about these before, but the simple concept is that x-rays are going to ionize the air inside of the ion chamber (i.e. free electrons) and then those ions are going to be collected in order to generate an electric signal. We get a measurement of an electric signal which is proportional to the x-ray exposure incident on the chamber.
We described the setup above and now we are ready to go into the half value layer (HVL) measurement itself. We first perform the measurement without anything in the beam. Just measure with the ion chamber itself, this gives us our reference (i.e. the exposure without any additional filter).
In the plot in the figure all the values will be normalized by this reference value, so the largest number will be 1.0.
After we have made the reference measurement, we insert a thin sheet of metal into the beam. Then we take a second exposure measurement. The measured exposure fraction is the second measurement divided by that exposure without a filter. For each additional thickness of material that is inserted we can make a measurement, and each measurement will generate one point on the curve in the figure.
After we have several points on this curve, we can do a curve fit. From the curve fit we can see the relationship between the material thickness and the exposure fraction. After we have done the curve fit we will be able to estimate the half value layer. Remember, the definition of the half value layer (HVL) is when the exposure is 1/2 of what it is without any material.
Even if we don’t have a measurement that’s exactly at the half value layer (i.e. the perfect thickness of materials to stop have the x-rays) we use curve fit to estimate the HVL. This is demonstrated in the figure. We draw a horizontal line at ½ of the exposure, then where that line intersects the curve we draw a vertical line. The intersection of this vertical line with the x-axis give the HVL. The HVL will be in distance units, e.g. mm.
These HVL measurements characterize that in our system, we can track that over time or we can compare different systems with that same measurement idea. So, that’s called the half value layer. As we have discussed the half value layer is dependent on the beam spectrum and the material.
We have a separate calculator on Beer’s law , which is the way that x-rays are attenuated by a given material:
where I is the beam intensity after passing through the material, I0 is the reference intensity without the filtration, μ is the attenuation coefficient for the material for this energy beam, and x is the thickness of material. We can divide both sides by the reference intensity to get:
Our goal here is to solve for x since that is the thickness of material. The next step is to take the natural logarithm of both sides since this will help us to cancel out the exponential on the right hand side:
Then we multiply both sides by -1:
Now, we can apply our special situation here where we want to solve for the thickness x that we call the HVL (half value layer) when I/I0=1/2 (the beam intensity is half of the input).
After using our calculator to take the natural log of ½ we get the final relationship for the half value layer, where both the HVL and μ depend on the beam quality (i.e. these are really averaged over all of the energies in the x-ray spectrum).
Thus far we have been discussing the concept of a single half value layer of a material. We can generalize this concept to include multiple half value layers where each additional half value layer the beam intensity is cut in half again.
If we use more material then instead of attenuating to where we get just half of the x rays coming out, we could get 1/4 of the X rays coming out. So, if we had two half value layers, we get 1/4 of the X rays coming out. If we had 3 half value layers, then we get (1/8) a little more than 12% of the X rays coming out. If we have four half value layers, we get (1/16) or about six percent.
The relative intensity passing through the beam filter is just 1/2HVL.This is the definition for multiple half value layers and you can see how this looks pictorially in this figure.
Why is there a correlation between beam quality and HVL?
The half value layer is indirectly measuring the beam quality since a harder beam will have a higher half value layer (HVL) and a softer beam will have a lower half value layer.
Harder x-ray spectrum will require more thickness of material as low energy x-rays are preferentially attenuated. Since a softer spectrum has more low energy photons, which get attenuated more easily, the half value layer will be lower for a soft spectrum.
An 80kVp beam (soft) and a 140kVp beam are filtered (shaded region in the figures represents spectrum after filtration) and a larger fraction of the photons are attenuated from a soft spectrum.
This correlation between beam quality and half value layer means that we can use half value layer as a surrogate or indirect measure of the beam quality. It is not a perfect metric but it does provide significant information about the spectrum and can be used for quality assurance trending. For instance if the HVL is suddenly reduced significantly on a system this means that there has been a change to x-ray spectrum which should be investigated.
Is there a correlation between beam quantity (mA) and HVL?
Knowing what you now know about HVL and beam quantity do you think that these two quantities would be related?
As we discussed above the beam quantity does not have an impact on the beam quality, i.e. an increase in the mA does not change the shape of the spectrum but only scales the spectrum. The half value layer will not change for these different acquisitions since the HVL measurements are all relative measures, so they do not depend on the mA. The main idea is that the half value layer will be conserved after the mA has changed, because the shape of those spectrum is staying the same.
Rad take home points:
- Half value layer (HVL) is measured experimentally by placing sheets of metal in the beam
- HVL is a surrogate measure of beam quality.
- HVL is higher for a harder beam and lower for a softer beam.
- HVL is not affected by the mA (i.e. the beam quantity).