The name is Gorithm... Al Gorithm

If you haven't been introduced to Mr. Al Gorithm (algorithm) yet, I can assure you he's not sipping a vodka martini - shaken, not stirred. No, an algorithm is what you are selecting on CR/DR systems in the pre-processing moment when you pick the body part and position that you are about to perform.

The algorithm is a pre-programmed set of mathematical codes used by your imaging software that is responsible for the image appearance after processing. It basically gives you a "model" histogram that the image should look like, and matches the actual histogram produced by the image to the pre-programmed one. The algorithm controls the brightness and contrast (gradient processing), edge enhancement and smoothing (frequency processing), and even histogram equalization. Appendix for Digital Acquisition and Display: ASRT. Brittain, Burns, Nethery, Smith.

Here is a clinical situation you may be familiar with: You go to shoot a cross-table cervical spine for a patient who is in a collar and lying on a backboard. You use plenty of technique that SHOULD provide you with adequate density and an appropriate scale of contrast, and you might even get an optimal S-number of Exposure Index, but you have trouble seeing the C7-T1 space. If you've been around a seasoned technologist, or if you really understand computers, you may learn that changing the algorithm in post-processing may actually allow you to visualize that space without any additional exposure to the patient. You know that choosing "axillary shoulder" or "cross-table hip" will allow visualization of the part, so you go right ahead and do so, patting yourself on the back because you've upheld the standards of ALARA and you've managed to even save time getting the patient off of the backboard without having to repeat.

While those goals are excellent, and I commend you for striving to achieve them, here's the problem: This is a short-term solution to a long-term problem. Let's say one year down the road, the films that you took are requested with a subpoena as evidence in court. The patient may have been involved in an MVA and possibly had some long-term neck injury complications and could be suing... yadda yadda yadda. They call upon a specialist in the field (probably a Radiologist) to interpret your films in court that look so pretty, but say "cross-table hip" on the exam type. The defense lawyer questions whether or not the Radiologic Technologist who performed the exam really took Anatomy and Physiology, because obviously, the x-ray (as supported by the Radiologist on the stand) is in fact, of a cervical spine - not a hip. The films are thrown out as evidence as an insubmissible legal document because of errors.

Mr. Gorithm has seemingly bitten you in the rear while masquerading as your friend all this time. The real question here is how can we avoid this unpleasant circumstance? In order to answer that, we need to know who programs your algorithms for your CR/DR equipment. Usually, when this equipment is purchased and installed, a representative from the manufacturer provides the initial calibration and setup of equipment. It might be a good idea to write into a sales contract for this person to come back at a future date to update and/or make any additional changes that may have been unforeseen at the initial time of setup. Some larger hospitals may have someone on staff who could, in this situation, create a new algorithm for "x-table c-spine" that resembles the histogram of a "x-table hip." Just be aware of your service contract and its limitations to any non-manufacturer personnel making these changes... you don't want to void your warranty or any part of your contract.

Ideally, each projection or exam type should have its own working/calibrated algorithm to process it under. It may cost your facility a little bit of money to initially set this up, but how much will it cost the facility (or your patients) in the long run?