Reporting standards in radiology

When reviewing radiologic imaging (film and digital) artifacts can be differentiated according to the cause. Obviously if the equipment is the culprit under normal operating parameters it will not be marketable until it is engineered out of the design. If the artifact is due to the patient from some clothing, or removal appliance then attempts to remove those objects will be included in the protocol of the particular test or equipment. Some artifacts will be due to unchangeable factors of the patient, ie implanted steel/ metals that induce the artifact. What typically is done for the latter artifacts is cancellation by software manipulation of the data points (similar to noise cancellation in audio systems or by fourrier analysis of the data points and removing the outlier artifact points of data) Many design improvements from first to fourth generation of CT scaners or Computerized Axial Tomography (CAT scan) were software speed improvements. The new CT scanners, pushed by the need for better faster and more detailed imaging so to compete with NMR scaners and more demands for non-invasive testing of Cardiac and other organ systems, has pushed the hardware tech. But this is off the subject of artifacts. In summary, native machine artifacts under normal operating systems need to designed out of the system(by pre implementation reporting or studies) and artifacts from environmental changes need to be removed by protocoL or software adjustment. Obviously once a product is marketed and is discovered to create a previously unkown artifact or even worse causing danger to the patient, then reporting by the end-user to the marketer and to regulatiry agency such as the FDA in the US (for over-sight protection for the public).(all which is also cross regulated for radiologic-biologicals by the Nuclear Regulatory as well as now HomeLand Security Agencies)

Thanks for the technical detail. I've read a few papers about artifact remediation techniques, and these papers were pretty clear that this is a central objective for radiologists, in general. It also sounds like you have a better system of oversight in the US.

Here in Canada, with public health care seriously underfunded for decades, some radiologists have to work with less than perfect equipment. So the question comes up, what if the machine you are using has an "irremediable artifact" - something you can't fix, but that leaves a significant "data gap" in your scans. How would you deal with that, when you're making the written report of the examination findings?

My own view is that an unremediated artifact should be noted in the written report. But this is not the case - not in reports that I've seen in this jurisdiction.

I have a couple of questions:

#1. Are we talking about images on a CT scanner?

#2. If it is a CT scanner image is the artifact on the scans a fixed artifact in terms

of location on the scanned image or is the artifact variable in location?

The reason for the second question is that fixed defects or artifacts can be worked around by repositioning the patient. However this can in some cases double the radiation exposure on a CT scanner study. For migrating or variable artifacts you are left with no predictable maneuvers and is indicative of more serious system mechanical problems.

For the sake of the discussion, lets take an example from a CT scanner. The artifact on the scans is a fixed artifact: a ring artifact with a "hyperdense" artifact at the center of it, affecting at least half a dozen of the images in the series.

In this case there was no workaround, the patient was positioned at the beginning, not repositioned. The "irremediable artifact" was simply ignored and never mentioned in the report.

I've also seen uncommented artifacts in a flat plate xray. The radiologist afterwards identified the artifacts as caused by "dirty trays". (In this case, I can sort of guess why you wouldn't want to mention it in the report! ).

So we have a well-known artifact that is irremediable, not mentioned in the report, and we also have casual unremediated artifacts (someone didn't clean the trays?) not mentioned in the report. Leading me to ask, why don't radiologists follow scientific reporting norms...

I must say that the "standard of care' and medico-legal atmosphere in the US and Canada in terms of radiologic study reports (based on what you just described) is very different. First, a physician in the US would never send another patient to a radiologic facility that either failed to report "inadequate x-ray studies" or only commented on what they could see as normal...Most ordering physicians, will also read the x-rays even before the radiologists issue a final report... and would not accept the "incomplete or inadequate studies"...In my experience a dirty film cassette will be "wet-read" by the radiologist and before the patient leaves the x-ray suite another x-ray would be repeated until an adequate study is performed....I must say that the characterization of the system you describe concerns me in what I can expect in the US with the coming health care reform......Reporting standards here (US) are established by the American College of Radiologists for the specialty and I will find you the specifics of the standards and give you a posting link..

I think the health care reform in the US would not easily lead to the kind of problems we have here - because care will still be available on the open market. As long as there is competition, there's pressure to perform. But in Canada we have a government monopoly on health care services. This is really a bad situation. It was one thing fifteen or twenty years ago when the system was reasonably well funded. But then the federal government decided to cut back on health care to balance the budget. This went on for years - now the system is badly crippled.

As for the standard of care, I'm wondering to what extent this is regional, or even more localized to a specific hospital or health authority.

The Canadian Association of Radiologists has a published communication standard which is available here. The reporting standard is scientific, but there's freedom to dispense with it: the problem seems to be, that the CAR has a disclaimer on all of its standards, including communication. The standard itself is clearly not met in the examples I gave you. It appears that the institution here has taken liberties with that waiver - and adopted a format and common reporting practice that is unscientific (including, using the term "Impressions" instead of conclusions, reading etc).

My feeling is, scientific reporting standards should never under any circumstances be subject to a waiver. I look forward to comparing the American standard.

Here are the American College of Radiologists standards both in communication and performance of various types of radiologic procedures. THey are called Practice Guidelines and will soon be a adopted by the World Heath Organization for inclusion in the United Nations World Health initiatives....may help with standardization world wide. Hope this helps its very well delineated.

http://www.acr.org/SecondaryMainMenuCategories/quality_safety/guidelines/dx.aspx

You are now the third Canadian that has given the same analysis of a a non-competitve single payor health care system. Your analysis however of the US continuing with a competitive free market alternative may only be short lived, I hope not..

Very interesting. They have exactly the same sort of disclaimer at the beginning, guidelines and not legal standards and so on. But perhaps in the US, the hospital or radiology clinic establishes its own standards? This sort of ad hoc doesn't bode well for quality assurance.

There's no scientific organization on earth that would preface their reporting standards with the freedom for any scientist to waive or alter or adopt some other standard in reporting.

Medicine is supposedly self-regulating, but the standard of reporting can be waived or altered without risking professional censure. Where is the regulation?

I notice that the use of "Impressions" instead of "Conclusions" is not an invention of our own jurisdiction, it's actually the American standard. This really erodes the scientific credibility of a radiology report.