Resolution of Definition Axial Datasets
The following question has been sent by Judy Oltmann, CTA Technologist, USA:
My question may have appeared vaque before. My radiology group has had a consistent complaint concerning the mottle or graininess, both terms applied, to the axial datasets acquired during our coronary scans with the Definition scanner. We have applied new kernal and slice thickness, increased flow rates for contrast and a fixed pitch with certain patients. The temporal resolution is too grainy for them at 82 ms, but much sharper than a reconstruction at 165 ms, resembling a 64 slice cardiac acquisition. The comment is that the 82 ms recon makes the diagnosis of a soft plaque difficult due to the mottle on the axial dataset.
I would like the opinion of other groups who are interpreting cardiac studies to give me some insight as too any possible solutions to their complaints. Also understand that in South Texas, our patients population averages 200 LBS and more.
Pal Suranyi, MD, PhD, Medical University of South Carolina, USA:
Dear Ms Oltmann,
Thank you for your enquiry and for visiting DSCT.com.
Quantum mottle, or graininess is most often caused by the following two things: lack of sufficient photons, or by the particular reconstruction kernel used. I would be interested to know which kernel you specifically tried and did not work for you.
As you rightly pointed out, larger patients “absorb” more photons, thus the detectors get less information from which the scanner can recontruct the images. Here at MUSC, for “generously proportioned” (quoted from Dr Schoepf) patients, we use thicker reconstruction slice (maybe 1mm instead of the 0.75mm that we use for routine patients) and an even smoother kernel (B10 instead of B25f), and increase the tube current slighlty. If the patients have calcifications and stents, we do two separate reconstructions, one with a sharper kernel (B46) to review calcium and stents, and a smoother reconstruction for reviewing soft plaques and native vessels.
In larger patients, giving a little more radiation does not necessarily mean that they will have a higher risk of cancer, because the “per volume”, the radiation exposure will still be the same.
I personally do not think that using a better temporal resolution in itself, i.e. 82ms vs 165ms, will increase the mottle. What I think may be possible is that you have less blurring due to motion and crisper contours, which, if you use a sharper kernel may translate into mottle. I think one of the greatest advantages of Dual Source CT is the outstanding temporal resolution that exceeds all other available scanners, and this is crucial in cardiac imaging and is worth exploiting.
In summary:
Try using a softer kernel e.g. B10 for reconstruction Increase mAs (the scanner will alert you if you go too high) Increase the reconstruction slice thickness to 1mm
I hope this helps, and please do not hesitate to let us know if we can be of further help.
Best regards,
Pal Suranyi , MD, PhD
See another answer to this question by Dr. Nicolaou
See another answer to this question by Dr. Achenbach
