Another CT dose related question

big man

Sorry for having to ask another question, but there is just one more thing that I want to get my head around.

Let's say you have this protocol:

Standard Resolution
64/40 x 0.625 collimation
2.5mm thickness
40/25 mm increment
240 scan angle
0.4s rotation
120 kv
55-165 mas
512 matrix
75% prospective phase
220 DFOV
approx 120 mm scan length
Filter CB
100 mAs

OK now here is my issue. I've been told that you can estimate effective dose by using the following formula:

[tex] Effective dose = DLP * k[/tex]

Now the k-factor is a normalised conversion factor and can be found at this site:

http://www.drs.dk/guidelines/ct/quality/mainindex.htm ------- Appendix 1 of Chapter 1

What I don't understand is the scan angle. I thought that the CTDI and therefore the DLP were given for 360 degree scans. So therefore your estimation of effective dose (if you only use a 240 degree scan angle) wouldn't be correct if you used the above formula.

For example, let's say that you performed 2 scans of the chest. For both scans you have the EXACT same protocol (same kVp and mAs) except for the scan angle. One has a scan angle of 180 degrees and one has a full 360 degree scan angle. I'm just not understanding how you can use the same formula for both these situations. I mean if you are scanning 180 degrees that is centred beneath the patient then you aren't uniformly irradiating all the organs in the chest region are you? I'm really confused about this because I just don't think that you can use that formula yet I've been told that the dose isn't dependent on the total scan angle?

I hope this hasn't been to muddled of an explanation, but I'd appreciate any assistance.

imabug

There's nothing in the definition of CTDI that references the scan angle. Scan angle just determines the number of projections used to reconstruct the slice.

Keep in mind that diagnostic dose estimates are just that, estimates. At best whatever number you end up with is only going to be accurate to within a factor of 2-3 at best. Most of the time, if I feel the dose estimates I calculate are within an order of magnitude, I'm happy with that.

Tsu

If you used a scan angle of 240 instead of 360, wouldn't you multiply your estimated DI by .66 (2/3) - because you are essentially cutting the amt of info you are collecting by approx .66?

imabug

if you're maintaining the same mAs per slice though, the mA needs to go up by 3/2 to compensate for the reduced scan time

Tsu

Does reducing the projections (scan angle) actually reduce the time - and visa versa?

I was under the impression that when I reduced my time, I was still getting 360 projections - just with faster rotation.

If you leave the mA the same, are you're saying reducing the scan angle would not be an acceptable method of reducing dose? Normally, I would just lower my mA or time to reduce the dose.

If you had a moving target (drunk) on the table would it be a good idea to up the mA and decrease the scan angle to have less chance of motion?

imabug

this is still true. As i'm sure you're aware, scan angle normally isn't a parameter that's available to be modified. On all the scanners I've played with, when time is a modifiable parameter it usually affects the tube rotation speed. However newer scanners can use a partial scan technique (partial reconstruction would probably be a more accurate term) for doing cardiac studies which enables them to capture heart motion and do retrospective cardiac gating. Reconstructing slices based on a limited number of projections is all handled by the recon algorithm. I've never seen it under user control on any of the scanners I've played with.

since the tube still has to rotate the full 360 deg and is probably still producing x-rays for the whole rotation, the total scan time won't be reduced and total exposure to the patient would remain the same. partial scanning just gives you the ability to reconstruct more slices per rotation

you could reduce the scan time (increase rotation speed) and use a slightly higher pitch to reduce the total scan time, or strap the patient in really tight. a partial scan technique could potentially reduce the amount of motion in the reconstructed slices.

Tsu

Thanks, imabug!

Moonbear

I realize this is a serious response, but it made me giggle a bit, because it's exactly the sort of answer I would have suggested knowing nothing else about operating the equipment.

Tsu

Strapping and taping is just a given in my job. Expecially if the patient has been drinking or appears to be on a recreational pharmaceutical of some kind.

big man

Thanks again for taking the time to explain this stuff to me Imabug.

So basically you are saying that if you have used a 240 scan angle it means that this is all the information you have used to reconstruct the slice, but you have not actually reduced dose. Even with prospective gating the tube is still "active" for the full 360 degree rotation then?

I think that my main problem is that I do not have a clear visualisation of the image reconstruction in a CT system. Can you recommend any good resources regarding CTs that might clearly show this at all?

imabug

yes. in practice, the x-ray tube is energized for the entire rotation. otherwise you're just wasting useful opportunities to collect additional data. Partial angle reconstruction is handled in the reconstruction algorithm and simply determines the number of projections used to reconstruct the slice.

you can certainly envision a CT scanner that does partial angle acquisitions by having the x-ray beam on for x degrees and then turning off for the remainding 360-x degrees. Then the radiation dose delivered depends on the constraints you set on kVp, mA and time. If you specify that the mAs is constant compared to a full 360 degree acquisition, then because the scan time is reduced mA needs to go up. Thus radiation dose remains the same.

go to CTSim.org and download the program. It's a CT simulation program that may help shed some light on the mechanics of CT acquisition and reconstruction.