A review of radiation dose measurements in CT and a calculator for changing multiple technical parameters (kVp, pitch, s) and observing the effect on the mA needed to keep the dose relatively constant. We also provide a couple use cases that would motivate changing the technical parameters of a protocol at the time of scanning. This is just the basics and does not cover advanced automated exposure control features where the scanners adjust the mA during the scans for improved dose efficiency.
CT Dose Dependence on Technical Parameters
As a technologist you are the primary user of modern CT scanners and the most common thing that you do on these scanners is adjust the technical parameters (e.g. the kVp, mA, rotation time). Modern CT scanners include automatic methods to estimate the radiation dose (or at least a surrogate of the radiation dose, CTDI). For any protocol that you are using the scanner can report the estimated CTDI that the system will deliver for those technical parameters.
What we would like to focus on here is an intuitive understanding so that you know which parameter has the largest impact on the radiation dose (for the same relative change in different parameters). We also want you to know which parameters are directly related to the dose (i.e. an increase in that parameter increases the dose), and which parameters are inversely related to the dose (i.e. an increase in that parameter decreases the dose).
Having this understanding will be very helpful when you are modifying technical parameters as you will be doing less guess and check on the fly at the console. Finally, we also describe an online calculator that will give you a rough estimate of how you would change the tube current, mA in order to keep the dose constant for a manual technique (i.e. not using the automated exposure control software).
Before we get to that though we will give a brief lesson on how radiation dose is measured and what parameters are reported on a CT system. For a more complete reference on radiation dose units and their relationship please refer to our Radiation Dose post.
CT mA Calculator in Action
We will start out by going through some example uses of this RAD calculator to build our confidence so that we can internalize these relationships and we will have more confidence of the relationships between the technical parameters on the scanner. After these examples we give more background on dose measurements in x-ray and CT and show the equation used in this calculator and describe it.
Effect on kVp:
Just to make the numbers nice, let’s say originally we have a tube current of 100 mA, and then originally we have a tube potential of 140 kVp, a rotation time of 1 sec and a pitch of 1.
In the calculator if we start off with 140 kVp in our parameters and our new tube current is 100 mA because we haven’t changed any of the parameters yet. Imagine we’re doing a scan where we want to go to very low kVp because we really want to bring out the iodine contrast. So we’re going from 140 kVp to 70 kVp. Originally we needed 100 mA. So what do you think we’re going to need for the mA if we go from 140 kVp down to 70 kVp?
We’re going to need about 400 mA because our new mA has to compensate for the fact that the kVp is the biggest driver and that goes approximately as the square of the kVp. Therefore, if we think of 140 over 70, that’s a factor of 2 and if we square that, that’s a factor of 4. So, now we need 4 times more mA on our system in compensate for the original higher flux at the higher kVp.
Effect on rotation time:
If we start with a rotation time of 1 second, and then we go to a 1/2 second rotation time, how much mA do you think we would need? We’re going to need to deposit energy more rapidly since we are spinning faster, if we want approximately the same dose to be delivered. If we have half the amount of time to acquire it, we’re going to need twice as much mA during that given time. So if we start with 100 mA and just change the rotation time to make the gantry spin twice as fast, we will need twice the mA or 200 mA.
Effect on helical pitch:
Then finally, we can think about pitch. In the scenario that we want to reduce the helical pitch from a pitch of 1.0 originally to a pitch of 0.5 what will the effect be on the mA required to keep the dose constant. If we started at 100 mA then we will only need 50 mA after we reduce the pitch. This is because now we’re essentially double scanning from the perspective of the pitch because we’re going from a pitch of 1 to a pitch of ½. Since we are acquiring essentially twice as much many views of data each view should only receive half of the original mA in order to keep the dose constant.
Background on CT Dose Measurement
Exposure Measurement in Air
If we want to make a measurement of the radiation exposure or the dose that’s deposited in the air, we have to use an ion chamber filled with air. There’s a voltage that’s applied across the chamber, you can think of it as positive and negative plates. When the x-rays are incident on the air within the chamber the x-rays ionize air molecules and electrons are produced (along with positive charged particles (ions) from the molecules that lost electrons). The electrons flowing through the air because of the electric field will be measured as a current in the circuit.
Once we can measure that current of electrons, if the ion chamber and the electronic circuit is calibrated, the software with the ion chamber will calculate the exposure which is measured in Roentgen (R).
Then what we really care about more for patients is measuring in a unit that is more similar to a radiation dose received by a patient, so something that’s more water-like rather than air-like.
The units of radiation dose are (Energy/Mass) and the SI unit used for radiation dose is the Gray (1 Joule/ 1kg). For a radiation dose of 1 Gray there is 1 Joule of energy deposited in each kilogram of mass.
To measure the dose rather than the exposure you can use that same kind of setup of having an ion chamber, but typically we’ll place the ion chamber inside of a more water like material so that it is more similar to the situation of a patient.
CTDI (Computed Tomography Dose Index)
In computed tomography, what we typically report on the system is called the CT Dose Index. The CTDI metric corresponds to the dose in plastic phantoms. The CTDI is not exactly the patient dose for a given exam since the patients are not exactly the same as the phatoms.
However, the CTDI is good for comparing protocols, scan techniques, or even different scanners. We use a plastic phantom to measure the CTDI and insert an ion chamber at different points in that phantom. The standard protocol uses 4 locations peripherally and one location in the center of the phantom. We measure the dose in the center region and then we do a weighted combination of 1/3 the dose in the center and 2/3 the dose peripherally to get the CTDI.
There are also two phantoms that are used: 16cm diameter (closer to heads or small bodies) and 32cm diameter (closer to standard and large bodies). To calculate the CTDI, we measure the calibrated dose in milliGray (mGy) at each location (i.e. 1/1000 of a Gray). The values of mGy are used because the doses from CT scans are typically on the order of several mGy, and we would be writing a lot more zeros if we reported everything in Gy. We measure how many milliGray peripherally, so in these outside regions.
Although we mentioned already, we just want to drive home the point that CTDI is not equivalent to the patient dose. We scan patients of all different sizes (e.g. x-ray attenuation values). The measurements of the CTDI phantoms correspond to the dose deposited in the plastic phantom rather than the dose deposited in an individual patient. The largest reason that the CTDI does not correspond to the measured dose is that for the same x-ray parameters the dose deposited is strongly dependent on the patient size.
There are proposals for future methods such as Size Specific Dose Index (SSDI), which will be closer to the patient dose as they will be normalized based on the size of the patient. However, SSDI is not yet implemented clinically on commercial scanners.
An example of the size dependence is if we use the same x-ray exposure (i.e. the same technical parameters like kVp, mA, s, and pitch), and we run the same exact same scanning except comparing the two different phantoms. For example we start with the larger phantom and then afterwards we switch to the smaller phantom. We would measure a radiation dose of about two times higher in the smaller phantom. This is because the x-rays are attenuated more in the larger phantom and thus don’t deposit as much dose as they travel deeper in the phantom.
Obviously, in clinical scanning the technical parameters are changed to account for the body habitus as the image quality of the CT scan would be much worse of the larger phantom at the same dose. This is just one example to demonstrate that there is a dependence of the radiation dose absorbed on the size of the object in the scanner.
However, even though the CTDI is not exactly the patient dose it is a useful measure to track the radiation dose and make comparisons between different technical parameters. Now that we have a dose surrogate, we next need to take into account how much of the patient is scanned along the SI direction (i.e. superior to inferior).
DLP (Dose Length Product)
We need a direct method to account for the length of scan as a scan that only covers a single 5mm needs to be distinguished from a scan that covers the chest-abdomen and pelvis and is 500mm. The CTDI just considers the dose in a given axial plane and this is extended to volumetric scanning using the Dose Length Product (DLP).
As its name suggests the dose length product is just the multiplication product of the radiation dose (mGy) x the distance scanned (cm). So the units of DLP will be mGy x cm.
Using this simple relationship you can see that if your scan covered twice as much length that the DLP would be twice as high, assuming that the other technical parameters were not changed.
The DLP is also reported automatically on CT systems and is useful to track the doses and properly account for the length of scanning.
This has really just been background on the dose measurement methods in CT as the primary motivation of this post is to discuss the relationship between the technical parameters on the system and the radiation dose.
Modifying Technical Parameters
This is just as a background for the way that radiation dose is measured on the CT system.
Next, we will discuss a few motivations for changing the technical parameters on the CT system, i.e. why don’t we scan all patients with the same technical parameters. Our primary motivation for modifying the technical parameters is to tailor the acquisition to the specific imaging scenario which includes the clinical task and characteristics of the patient.
For instance, if the patient is a young pediatric patient without sedation one of the biggest concerns could be the possibility of motion artifacts degrading the image quality. In this case a fast scan is beneficial, so you potentially will be using a faster rotation time and a higher helical pitch.
Also, one of the most important parameters for many clinical scenarios is the contrast enhancement within the soft tissues. In this case there is the desire to use a lower kVp in order to improve the visualization of iodinated contrast for instance. The noise in the CT images will also increase but the image contrast for iodine will increase faster than the noise. Therefore, for imaging tasks which are dominated by iodine contrast there is a motivation to reduce the kVp, as long as you can still achieve sufficient flux (i.e. so that enough x-rays make it through the body to be measured on the detector). This is another example of a parameter that is often changed from patient to patient, and CT manufacturers even have methods to assist the user to select the optimal kVp for each patient.
These are just some reasons why you would want to modify the technical parameters at scan time rather than use the same protocol from all patients. On modern systems much of the time you will be using an automated exposure control system so that you are not directly setting the mA level. But, if you have modifying a manual technique (i.e. a protocol without automated exposure control) you will be setting the mA (or equivalently mAs if your vendor doesn’t let you change the mA directly).
Dose Relationship on Technical Parameters
Regardless of if you are using a manual technique or an automated mA prescription it is important to have an intuitive feeling at how the different technical parameters that are changing will effect the delivered dose. The major technical parameters that we will discuss are: tube current (mA), rotation time (s), tube potential (kVp) and helical pitch.
The rad calculator that we have here is for the question, If I change some of the technical parameters on the system and I want to keep the dose roughly the same how I should change the mA. First we provide a table with the major technical parameters that you can change on the system and how they effect the radiation dose.
|Effect on the radiation dose|
|mA – directly proportional|
|kVp2 – power law approx. square|
|s – directly proportional|
|1/pitch – inversely proportional|
We note that the relationship with kVp is not exactly squared (i.e. a power law where the power is between 2-3), but it will be easier to use the square here as an approximation. The most important thing to remember is that changes in the kVp have the strongest effects on the radiation dose.
We also note there are other factors such as the collimation and the scan field of view (i.e. the selection of the pre-patient collimator). There are other factors involved in the radiation dose, but we just want to really have a sense for these dominant factors and how they relate to the radiation dose.
Changing mA to keep Dose constant
Imagine we’re doing a manual mA scan and our original scan parameters had an mA, a kVp, a rotation time, and a pitch. Then imagine what we want to do is we want to prescribe a new scan with approximately the same radiation dose. We can write an equation to solve for the mAnew assuming that we know the other new parameters. In this figure we show how to write out the equation with the left hand side being the new parameters, and the right hand side being the original parameters (in your default protocol) for instance.
If we solve this equation for a new mA, our new kV will come down into the denominator. Our new rotation time will come down into the denominator and our new pitch will go into the numerator. So the relationship is that our new mA is proportional to the original mA x (the original kVp divided by the new kVp)2 x (the original rotation time divided by the new rotation time) x (the new pitch divided by the original pitch).
This is the relationship that we have implemented in this RAD calculator so that you can practice this so that when you change the pitch on a protocol you won’t have to consciously think about it as you will have the relationship down.