Technical Issues in Digital Imaging: How Do They REALLY Impact Clinical Care?

Dale A. Miles DDS, MS, FRCD

Forward to Part II

It's truly hard for the practising dentists to begin to sort out the "hype" from the truth about digital x-ray imaging systems. How many line pairs/millimeter is good enough? Why do I need 4096 gray levels? Why are some systems "wireless"? And, how do they do that? What is the difference between a CCD and a CMOS receptor? Do I really care? All of these questions arise each time I address a dental group. Then there's the ultimate question: how will digital imaging improve my practice and my patient's care?

I will attempt to answer all these questions in this two-part article. In this first part I'll discuss hardware; that is, the sensors (receptors). In the second article, I'll present information on the software or "image processing", the part of the system which truly impacts patient care.

The "hardware"

You can choose from 3 direct digital receptors: the CCD (charge-coupled device), the CMOS (complimentary metal-oxide semiconductor) or the PSP (photostimulable phosphor plate). All are quite capable of giving you a diagnostically accurate, reproducible image for any of the routine dental tasks: caries assessment, periodontal bone level information or periapical lesion detection. You can read more about the principles of all of these receptors using the references at the end of the paper." Buyer's Guide 2001 " details a comparison of the current commercial systems.

To begin, let's talk about some technical specifications of the hardware, and the claims by which the manufacturers try to win you over.

Line pairs/mm. is a measure of the system's ability to capture detail; that is resolution. Manufacturers claim anywhere from 6 to as much as 22 lp/mm. Does it really matter? In truth the human visual system can only resolve about 8 lp/mm at the best of times without magnification. So does 12 lp/mm or 20 lp/mm resolution really matter? The answer is NO! It doesn't matter…all of the current digital x-ray systems use sensors capable of capturing a diagnostically acceptable image.

4096 grays are the number of gray shades a CCD or CMOS receptor is capable of displaying on a computer monitor that captures a 12-bit image (this is just the number 2 raised to the 12^th power - do the math!). Some CCD or CMOS receptors actually capture this amount of gray level information. BUT, none of the manufacturers actually display a 12-bit image because the computer monitor for all of these systems is usually capable of displaying ONLY AN 8-BIT IMAGE - or 256 shades of gray! So, do you need 4096 grays or more (PSP systems have even more gray levels or shades). The answer is NO! It doesn't matter. And, even if you bought a very expensive medical grade monitor to display all those grays, the human visual system can only separate the differences between 64 grays! Once again…it doesn't matter!

"Wired" vs. "Wireless" The PSP systems are "wireless". That's because they use plastic plates coated with a phosphor material sensitive to x rays. The phosphor material stores the x-ray energy, like a latent image with film, until it is scanned by a laser. The light excited and released by the laser is captured as an electrical signal which is then converted to a digital image (analog to digital conversion) for display on the computer. The image is displayed continuously by the computer's "frame grabber"). The image is thus "digital" and can be electronically processed just like the CCD or CMOS image. However, the PSP systems are slower because the plate must be placed in a drum scanner (DenOptix or Digident) or in a slot in the computer for laser scanning (Digora) in this 2-step fashion. Thus the CCD or CMOS systems are faster. The image appears on the monitor within seconds after exposure. Retakes then, if necessary, can be performed more quickly. You must decide if the saving in "tech time" is worth the perceived "hassle" of having a wire. Most clinicians and auxiliaries I've talked to say the wire is not a problem…you just have to use the positioning instruments provided with the system. Which should I choose? Here it matters! But, it's largely a matter of personal preference. Remember, all of the systems improve your imaging capability and all are "up to the task".

CCD vs. CMOS This is a controversial area since CMOS receptors have only just become available for x-ray use. They are great for video and digital still cameras, video game interfaces, etc…. They use much less power, are cheaper to make because they are used in EVERY computer already (without the imaging sensor portion), and work very well in bright light conditions. However, they have not been tested for x-ray image acquisition yet, and have more noise (electronic cross-talk that degrades an image), are less efficient at gathering light and x rays, and thus have a lower quantum efficiency than CCDs. This means they gather less x or light photon information and thus may not have as much diagnostic information to display. The CMOS chip has a microprocessor on it, some RAM (random access memory) as well as the sensor/receptor. This means that there is less imaging "real estate" for the device, and thus maybe less capability of recording incoming x-ray information than the CCD. CCDs for x-ray imaging are a very stable and mature industry. They just cost more. In order to compensate for less x-ray gathering, the CMOS sensors have microlenses and scintillators (like CCSs) bonded to them to gather more light. This "optical packaging" adds to the cost of the sensor. So where are the savings? I don't know, and we probably won't see any! By the way CCD is not really "older technology" as claimed by some manufacturers. It was better technology for x-ray imaging, and may still be. Table 2. Details the comparison between CCD and CMOS features. So, CCD or CMOS? Here again it matters…for my money, I'd stick with CCD for now, and wait until the evidence from the academic/research community is in.

Table 2 - CCD - CMOS RECEPTOR COMPARISON
Note- Bold Words are an advantage of the system.