Current Clinical Concerns in CT: Results : CT Scanner Hardware
Questions pertaining to number of detector rows
1. Our regional hospital serving a population of 60,000 is purchasing a new multislice CT. MRI and angio are located in a neighboring health district, approximately 1 hour away. Our medical staff that includes internists (cardiologists), general surgeons, orthopedics and urologists. We feel 16 slice CT is most appropriate scanner for the short and long term needs of the health region. It would allow use to best offer cardiac CT, CT angio, etc. Do you agree, or are we buying more than we will need. Our experience for the past 4 years has been with a single slice helical scanner.
| Answer: I think 16-slice is ideal for your situation. 16-slice provides many of the capabilities that you need, but because of manpower, etc, have a hard time to provide. CTA can replace 80-90% of all diagnostic angiography and will become more valuable in cardiac imaging and peripheral vascular disease. You will be able to provide 1 stop shopping for many of your patients. If you buy, say a 4-slice, you will limit your options and prevent the growth of your practice. Call with any other questions or post another on www.ctisus.com. |
2. I don't see protocols for 8 row MDCTs. Is my group already behind the times?
| Answer: You are getting there. Most people have 4 or 16 slice. We will try to get some 8 slice protocols up, but soon we will be putting up 32 slice protocols. |
3. We soon will be operating a Siemens 6 detector scanner. Do you think this machine will be able to do coronary CTA?
| Answer: Interesting question. We currently do all our coronary CTA's on a 16-slice MDCT, but a lot of the original work was done on a 4 slice MDCT, so one should assume you could get by with a 6 slice MDCT. Let me ask around, and will report back to you. |
4. I would really appreciate your opinion. We have a Siemens Sensation 16 scanner, and whilst most imaging is good, there are some areas of extreme disappointment. In particular, small orthopedic work (including facial bones). The detail is just not there (fuzzy). Have taken to doing sequential scans on the 0.6 mm collimation but (as there are only 2 channels of this) the scan times are very long. Also, spine imaging is very poor. Very noisy at doses that should be OK. I read a response that suggested the ASA filter, but this seems like a band-aid rather than a cure. Siemens tell that these issues will be fixed with a software upgrade. Have you noticed an improvement in image quality with a VA70? I believe the Sensation 4 produced MUCH nicer imaging in these areas.
| Answer: The key to bone imaging is to use different reconstruction kernels. Sometimes the high resolution kernal makes the images too noisy when you use the 80/90 kernals. We use 50/60 for the 3D for this reason. Using Inspace and presets the 3Ds are very nice. As for the new software, changes are always ongoing. The newest software has less issue with metal for examples and various presets create a range of bone visualizations. |
5. Why is it that, even the most authoritative multichannel CT users still believe (and publish the "fact") that the number of rows of detectors determines the number of "slices" that one can obtain per second, i.e. a 4 channel CT has 4 rows of detectors, a 16 channel CT has 16 rows of detectors. This is clearly not the case. Please clarify.
| Answer: Here are some good references to help clarify things: |
References
Rydberg J, Buckwalter KA, Caldemyer KS et al. Multisection CT: scanning techniques and clinical applications. Radiographics 2000; 20: 1787-1806.
- Summary: This Radiographics article reviews MDCT technology. Detector rows, detector array design and acquisition protocol parameters are discussed. In addition, the detector design differences among various manufacturers are reviewed. This is an excellent source of technical information for those practicing multislice CT.