Our experience at UW Medicine Radiology mirrors that of the authors in this article. DECT image quality is very much better with the current reconstruction software. It now rivals SECT in image quality and is the same in radiation dose. But tissue characterization is better and iodine contrast is much brighter – you may need much less injected contrast (up to 70% less).
This interesting article documents both the degree of CT dose reduction from model-based iterative reconstruction and improvement in image quality when looking at lung parenchyma detail.
This study further validates that model-based iterative reconstruction can decrease CT exam dose by 50-80% without compromising diagnostic power. There now is a substantial accumulation of published reports of this type in multiple body areas and organs. The same is becoming true for blended types of adaptive statistical plus model based (minus the optical components) iterative reconstruction (such as ASIR-V).
This article highlights the wide variation in CT patient radiation dose between similar institutions for similar exams. Recent analysis of ACR dose registry data also suggests there is wide variation amongst different regions of the country.
Such variations suggest that attention to the details of CT technique and technology can produce CT exams at much lower dose – presumably without compromising diagnostic power.
This article illustrates how iterative reconstruction can be used to markedly lower CT radiation dose without significant impact on diagnostic content in CT exams.
For patients with Crohn’s disease who likely will have multiple CT exams over time, lowering dose is especially important.
Study concludes that ultralow-dose CT may substitute for standard-dose CT in some COPD patients
There are at least three different generations of iterative reconstruction, all of which enable substantial CT dose reductions without compromise of diagnostic power. While earlier versions of IR yielded 30% dose reductions, those with model-based IR or some blend thereof can result in 50-80% patient radiation dose reductions – with even better spatial and low contrast resolution. Access the full article on this study.
As this article demonstrates, iterative reconstruction is a very powerful way to reduce dose without impacting diagnostic ability. Key points of the authors include, “To reduce patient and operator radiation dose involves optimization of medical imaging equipment and best control of the equipment by the operator. … The results of our study confirm in a large patient number reflecting the routine clinical setting that the image noise reduction technology allows a significant reduction in radiation dose. … The substantially lower radiation dosage achieved in a routine clinical setting with the image noise reduction technique, provide further evidence of the substantial impact of the new technology. They indicate potential reduction in radiation dosage in invasive and interventional cardiology with more diffusion of newer radiation technology in clinical practice.”
All iterative reconstruction techniques powerfully reduce CT radiation dose in the 40-80% range – without compromising diagnostic power. And they all continue to be refined and to evolve, as this article illustrates. While the “look” of CT images may change from the noise removal, the diagnostic power is not compromised despite the substantial dose reduction. As radiologists, working with change is our future. The old days of nothing but filtered back projection are in our history but not in our future.
This article pretty well confirms what many have felt: model-based iterative reconstruction (MBIR) lowers radiation dose by 70-80% compared to adaptive statistical iterative reconstruction (ASIR), without loss of diagnostic power/information. While the images do indeed look different because there is much less noise and because of a slightly different pattern in the remaining noise, all the findings are there. Further, the anatomy and the findings are displayed as well or better.
So, in a young patient (under age 45) – especially if they are likely to be getting multiple exams – use of model-based iterative reconstruction is well worth the longer reconstruction time.
(To read more about CT enterography, Radiologyinfo.org is a great resource for patients.)
Paying attention to limiting Z axis coverage yields big dose saving dividends! See this article for results of this study designed to assess the safety and efficacy of radiation dose reduction in hospitals lacking iterative reconstruction.
This comprehensive article demonstrates the importance of CT dose monitoring and utilizing strategies to achieve ALARA (as low as reasonably achievable) doses while maintaining image quality for optimal clinical diagnosis. The authors also describe how the use of technology can improve the radiation dose efficiency of CT scanners.
This article highlights that it is possible to achieve much lower radiation dose CT scans for commonly employed types of CT studies – the CT for urinary tract stones is one of the most common.
While not done everywhere, attention to detail can produce remarkable reductions in patient radiation without compromising diagnostic power.
Use of a lower kVp will actually make stones a bit brighter.
Careful attention to patient centering in the gantry can make a difference of up to 40% in dose.
And the use of iterative reconstruction techniques is now widely accepted to not compromise detection, yet with marked dose reduction – whether it be statistical iterative reconstruction, model based iterative reconstruction, or some blend of the two.
Radiologists and technologists both need to understand the importance of these tricks and the physics behind each.
This interesting paper talks about the use of iterative reconstruction to help lower the radiation dose of screening CT colonography.
Of course, as with all screening exams, the first order of priorities is to do no harm – hence the motivation to keep the radiation dose especially low.
The challenge is to lower dose without compromising diagnostic power.
For about the past two years, here at UW Medicine (Seattle) we have been using Model Based Iterative Reconstruction (VEO, GE Healthcare) for all our CT colonography exams. As recommended in this article, we also keep the kVp low – 80 or 100, which also helps to reduce the dose.
The result is a very low dose exam, but with excellent image quality and low image noise. This helps to make great coronal/sagittal reconstructions plus very nice 3D fly-through on the post-processing workstation.
This article illustrates that Radiologists’ perceptions of image quality and content change as they become accustomed – over time – to the different noise pattern of the various types of iterative reconstruction.
In fact, no spatial resolution or low contrast resolution is lost with iterative reconstruction techniques – and diagnostic power is maintained.
Our work here at UW Medicine agrees with this report.
And it is important to know this because iterative reconstruction can result in 30%-60% dose reduction for all types of CT, without loss of diagnostic power.
Guest blog by Kalpana M. Kanal, PhD, Director of Diagnostic Physics Section and Associate Professor in the Department of Radiology at University of Washington
How low can we go in radiation dose without affecting diagnostic confidence for detection of low-contrast liver lesions?
In a recent article we published, we studied the impact of incremental increases in CT image noise on detection of low-contrast hypodense liver lesions. Clinical CT liver exams were obtained on a 64-slice CT scanner using automatic tube current modulation at a routine clinical noise index 15. An artificial image noise addition tool was used to increase the noise level in clinical liver CT images to simulate 75% (NI 17.4), 50% (NI 21.2), and 25% patient radiation dose (NI 29.7) scanning relative to the original images (NI 15.0; 100% dose). The images were reviewed by radiologists of varying experience who subjectively scored lesion detectability on all the images, original and simulated.
We concluded that there is little loss of detection sensitivity for low-contrast liver lesion detectability of CT exams scanned with a NI at least up to 21.2 compared to a NI of 15, a patient radiation dose reduction of 50%. No significant degradation was observed when reader performance was evaluated as a function of lesion size (>10 mm) and contrast (>60 HU) at 90% sensitivity. When lesion size dropped to <10 mm or contrast was <60 HU, sensitivity did drop to 85%.
This study had some limitations, the most important of which was that this study was a simulation and not a true study of CT scanning at lower radiation dose compared to high dose scanning which would have involved scanning patients multiple times. Nevertheless, this study was important as it demonstrated that dose could be reduced by 50% without affecting diagnostic confidence for detecting low-contrast liver lesions.
This very wise philosophy for implementing iterative dose reduction in any CT program was well presented at the recent MDCT meeting of the ISCT in San Francisco in June. A key component is to have regular and measurable ways for radiologists to regularly grade or score image quality as dose is ramped down slowly with increasing amounts of iterative reconstruction. With Model Based Iterative Reconstruction (MBIR), it may be possible to drop dose up to 60% compared to otherwise low dose adaptive statistical iterative reconstruction methods (ASIR) – but not in one jump. It takes time to get accustomed to the slightly different look of images with iterative reconstruction.
At least a month’s worth of experience should accrue before passing judgment on image quality. It is also important to guard against anecdotal cases used to render judgments, so experience over time is important. But with a methodical approach, a lot of progress can be achieved in overall dose reduction.
At the 2014 ISCT-sponsored MDCT meeting in San Francisco – dose reduction was a key theme during all four days.
Iterative reconstruction was a common theme of an overall dose reduction program. While adaptive statistical iterative reconstruction (ASIR) now has been well-shown to reduce average doses by up to 40% without impact on image quality, the hot topic was model-based iterative reconstruction (MBIR) in its various forms.
Consensus is now developing around MBIR being capable of 50-70% dose reductions incremental to adaptive statistical iterations. While image appearance may be somewhat different from that of filtered back projection, it is now pretty clear that such different appearance does not compromise diagnostic power. Indeed, with experience, some radiologists have developed a preference for the image appearance of MBIR.
As explained in this study, here’s another trick for reducing both the patient radiation dose and the patient iodine dose in cardiac CTA: lower the kVp to 100 or 80 or even lower.
Of course, you can accomplish this same outcome by using dual energy CT and viewing the vessels with lower keV or kVp while viewing everything else at higher energies.
CT to search for urinary tract stone is a very commonly performed procedure because both negative and positive results may have significant impact on subsequent patient care. Often the patients are younger since stones can occur at any age.
This article presents very encouraging news about significantly lowering the dose of a CT for urinary stones by using statistical iterative reconstruction – yet with acceptable image quality and no loss of diagnostic power.
This report adds to a rapidly growing body of data about both statistical iterative reconstruction and model based iterative reconstruction for various types of CT exams. This body of data almost uniformly reports substantial patient radiation dose reduction in the 30% to 60% range with equal or even better image quality.
At the recent International Society for Computed Tomography (ISCT) annual meeting, held in San Francisco, Dr. Eliot Siegel, from the University of Maryland, discussed an evolving technique for CT dose-reduction. This application does not focus on simulated image noise as a measure of image quality, but instead, works to more accurately depict that noise in low-dose CT scans. In other words, this technique aims to optimize dose based on what the radiologist needs to see.
According to the presentation, the future of low- dose optimization will rely on a combination of the visual perception system and sophisticated mathematical models designed to minimize the dose for every imaging exam without impairing the quality of the image to the radiologist. In fact, the current methods for low-dose optimization are already on the way out as awareness for radiation dose continues to evolve.
Dr. Siegel’s analysis of noise in the new world of iterative reconstruction is very sophisticated and thoughtful. Concepts of pink and white noise plus just-noticeable differences really are cutting edge. These ideas will clearly advance our understanding of how to get a world of fully automated dose minimization.
Dr. Siegel and his team of researches, as usual, are thinking profoundly and in very innovative ways. These concepts certainly raise great hope for a much more systematic future combined with much more sophisticated math to make patient doses even lower than we had previously dreamed! It will be an exciting next three years for dose reduction techniques and technology!
At a recent Society for Pediatric Radiology (SPR) meeting in San Francisco, one presentation addressed low-dose CT’s “bright future, but troubled present.” Texas Children’s Hospital’s Dr. R. Paul Guillerman touched on the many uncertainties and challenges involved in low-dose radiation optimization, citing that these goals are so complex that they may nearly be impossible.
But, let’s take another look at this. Yes—dose reduction is complex and full of potential traps leading to poor technique or image quality. However, that is not a reason to avoid implementing a dose reduction (technique optimization) program at your institution!
So, how do you lead your institution down a road towards dose reduction? First, start with Google. Read what you can find on the topic. Then, go to meetings, talks, and presentations given by experts with considerable experience in the endeavor. Finally, start at your own institution.
I would suggest trying one variable at a time. Implement a weight based (or cross-sectional area based) kVp selection program. Then, embark on understanding how to use weight (or BMI) based selection of Noise Index for automated tube current modulation. Don’t forget to check out the easy stuff – like patient centering, use of bismuth shields, and limiting Z axis. Realize that even your contrast injection protocols – volume, rate, blending – might benefit from a weight based approach. Decide how to implement iterative reconstruction, varied by body region.
With these steps you are set to embark on a continual journey. Get started… today!
At the recent International Society for Computed Tomography (ISCT) meeting in San Francisco, studies were presented showing that CT scanning with a new algorithm, called model-based iterative reconstruction (MBIR), could offer better image quality and lower radiation dose than scanning with an adaptive statistical iterative reconstruction (ASIR).
According to AuntMinnie.com, “researchers claim that MBIR outperforms previous efforts to maximize the utility of low-dose CT exams, with researchers reporting excellent image quality and enhanced lesion conspicuity.”
University of Washington is one of the international sites for the multi-center trial of MBIR. My colleague, Paul Kinahan, was one of the two scientists who reported on MBIR images at the ISCT meeting.
We are very impressed with the technique here – it may someday result in a further huge dose reduction for CT. At this point it is in early stages of development and assessment of its clinical impact has not yet begun. But it looks very promising!