A Primer on Intraoral Direct Digital Radiography—Part II: System Components
Martin D. Levin, DMD; Shelly L. Lee, DDS, MS; and John M. Sturgeon, MD
In Part I of this article, the authors discussed the advantages and disadvantages of direct digital radiography (DDR). In Part II, the authors will review various system components.
Modern x-ray generators are an important part of any digital system. Because direct digital detectors require less radiation than film-based systems, digital images are optimized if taken with an x-ray generator with an extremely accurate timer. Also, low milliampere (mA) and kilovolt (kV) characteristics and a high frequency direct current (DC) circuit will enhance image quality. Ideally, rectangular columniation coupled with a small focal spot will bring an x-ray generator into compliance with the National Council on Radiation Protection and Measurements Report No. 145.1
At this time, there are approximately 14 solid-state detector systems available for purchase.2 The direct detector technology is further divided into two main categories: the charged coupled device (CCD) and the complementary metal oxide semiconductor-active pixel sensor (CMOS-APS).
The technical infrastructure required to deliver digital images to the practitioner can be divided into a series of processes that are called “the digital imaging chain.” In actual practice, the digital imaging chain is not strictly linear because of side branches and recurring steps and loops, but for purposes of simplicity, it will be presented here as linear. Comparatively, the conventional imaging chain is more complex, involving the x-ray generator and the patient, as well as the film, the film development process, and the viewing and preservation of the film. Every component in the chain degrades the image to some extent, and the weakest link in the chain will provide the most degradation. Because the imaging components are in a series, the degradations introduced by each step result in significant losses. In the digital imaging chain, however, the digital detector has the most potential to capture and retain more of the original image information, because it eliminates many of the conventional components most subject to error and loss of image quality.
The digital detector is the key to image fidelity, because any information lost there cannot be recovered by even the most sophisticated software algorithms. Performance of the detector is central to image quality as seen by the end-user, and it is dependent on a large number of complex parameters. However, the classic measures of image quality (IQ) do not provide an overall gauge of image output quality to the visual observer. Although computer monitor selection, ambient lighting,3 and image compression all affect IQ, the most compelling issue when it comes to purchasing a digital radiographic system is the IQ of dental structures as perceived by the clinician.
It may be tempting to make a simple comparison of sensor systems by using limiting spatial resolution (LSR), commonly measured in line pairs per millimeter (lp/mm), and defined as the spatial frequency at which the observer can no longer detect a high-contrast test pattern under laboratory conditions. LSR is problematic, however, because it does not take into account other characteristics that affect the IQ. If one sensor is rated at 13 lp/mm and another is rated at 22 lp/mm, one might simply choose the sensor with the most lp/mm; however, the amount of signal captured per pixel is reduced as pixel size gets smaller, although the amount of noise may remain relatively constant for each system.
Consequently, the practitioner should be cautious not to overestimate the importance of LSR by thinking that higher resolution, as expressed in lp/mm, facilitates better detection of low-contrast objects. Increasing the number of pixels can result in a lower signal-to-noise ratio (SNR); as lower SNR produces lower IQ at each pixel, small-object detection may be limited. Compensating for poor SNR is possible, to a point, with increased radiation, but increasing dose negates one of the key advantages of digital systems. Moreover, excessive dosage can lead to detector saturation and reduced image contrast. Contrast performance is the ability of the system to display the actual contrast of an object, and the ability to make window and level adjustments is potentially one of the true benefits of digital radiography over film. Digital detectors generally have wide dynamic range with thousands of shades of gray, so they might depict areas that might be over- or underexposed on film. It should be remembered, however, that the typical computer monitor displays only 8-bit information; ie, a maximum of 256 gray levels. Window and level adjustment is needed to display all of the subtle contrast variations in a digital image, just as a bright light might be used to evaluate an overly dense analog film’s radiographic detail.
Rather than evaluating a host of interactive parameters (ie, signal-to-noise ratio, line pair resolution, etc) to judge the quality of a detector, a more useful characterization of image output quality can be expressed as detective quantum efficiency (DQE).4 DQE is the measure of noise and contrast expressed as a function of object detail. A DQE of 100% is a perfect detector, meaning all information that reaches the detector is transferred out of the detector. Therefore, the combination of very low noise and high contrast offers the best detection of low-contrast objects. Detectors with high DQE can provide the practitioner with a better chance of detecting a low-contrast object than film with a higher LSR. According to Farman (Farman A, personal communication, August 20, 2004), “DQE is useful when it comes to accurate measurement for endodontic purposes.” Despite this, Farman reports that DQE has not been used in many dental studies, because it is relatively difficult and time-consuming to perform. To this end, most experts agree that any single objective measurement should not be relied on to help make a purchase decision. Compare as many technical specifications as are available, along with actual anatomical images.
X-rays are outside of the visible light spectrum; thus, they must be converted to light to be recorded on digital detectors. X-rays are scattered at two points in the process: by the tissues they penetrate and by a scintillator screen. Scintillators are bonded to the detector and convert x-ray photons into light. Such light is then detected by the CCD or surface for processing and depiction at the presentation tier. Scintillation layers are made of materials that include phosphor deposition Gd2O2S:Tb and columnar CsI:TI, where a Cesium coating is grown directly on the fiber optic plate and subsequently bonded to the detector.
CCD and CMOS-APS sensors use two different technologies for digital image capture, each with unique strengths and weaknesses that make them useful for different applications. In a CCD sensor, every pixel (ie, picture element) is arranged in a matrix of output nodes, which convert electron packets to voltage that is buffered and sent off-chip as an analog signal. The CCD detectors require many supporting components, such as timing generators and shuttering and signal processing chips, each located separately from the detector. The CMOS-APS detector, on the other hand, is a sensor where each pixel produces its own charge-to-voltage conversion, and the sensor includes many of the necessary digitization circuits onboard, so the output is digital. The CMOS-APS sensor also requires less power than a CCD detector by a factor of 100, which may improve the reliability of the sensor.5 Simply stated, a CMOS-APS chip uses considerably less power and may be more reliable.6,7 As of 2008, the most widely used wireless direct digital detector is the CDR Wireless™ (Schick Technologies, Inc, Long Island City, NY). The Schick wireless sensor is a CMOS-APS detector, which transmits the image data by using a 2.4 GHz band radio frequency signal to a remote receiver mounted in the operatory. The wireless and wired versions of the Schick sensors produce identical image quality. While most CCD and CMOS-APS detectors are wired and connect to the computer, PSP plates are more similar to film. They are thin and bend slightly, so they may be easier to place in more challenging cases, and a single centrally located scanner may service all of the operatories in an office, whereas one DDR system may be required for each operatory where radiographs are required.
A study by Yoshiura et al has demonstrated that most direct digital radiographic technologies, if optimized for contrast, exceed film-based imaging systems in the detection of small mass changes (holes created in a test phantom).8 Paurazas et al5 compared the diagnostic accuracy of CCD and CMOS sensors with E-speed film in the detection of periapical lesions created in the cortical and trabecular bone of 10 dried human mandibles. The images were evaluated by six endodontists and two radiologists. They concluded that CCD, CMOS, and film systems performed equally, and the digital systems required less radiation to obtain equivalent diagnostic information. Consistent with earlier studies by Bender and Seltzer,9 detection was significantly more accurate when the lesion involved cortical bone as opposed to just trabecular bone. Nair et al10 found that when E-speed film was compared by five observers to raw and enhanced digital images to evaluate crestal bone with tissue-equivalent human skull phantoms, film was not significantly different from the digital radiographs. Although these studies were in vitro and subject to errors inherent in artificially created lesions, Farman et al11 compared E-speed film and digital imaging to the actual measurements taken from periradicular surgical sites using impression material. Fourteen examiners assessed 28 lesions, with 10 randomly reread after 2 weeks. The detector images were judged superior to film with regard to accuracy in the measurement of periapical lesion dimensions.
Caries detection remains one of the most significant diagnostic activities in modern dental practice, with noncavitated carious lesions in the initial stages the most challenging to detect.12 Wenzel13 performed a literature review and stated that “digital intraoral radiographic systems seem to be as accurate as the currently available dental films” for detection of caries. In another review article, Wenzel14 stated that although digital detectors were as accurate as film for caries detection when a good quality image was acquired, placement errors increased with digital systems. Although radiolucency in dentin is a good predictor of decalcification, radiography is not a good predictor of initial caries in enamel occlusal and proximal lesions. In an in vitro study, PSP systems were judged by 10 observers as equivalent to E-speed film, and it was found that magnification of the PSP images significantly improved diagnostic accuracy.15
To determine the accuracy of film-based images and digital detectors, Lamus et al16 used the CDR™ (Schick Technologies, Inc) segmental measurement tool and conventional film measurement techniques. When compared to the actual file length in extracted teeth of different shapes, both film and digital systems were comparable in accuracy. Although direct digital images tended to overestimate the working length, E-speed film resulted in underestimating the tooth length. Research by Ong and Ford,17 Mentes and Gencoglu,18 and others have consistently reported equivalent efficacy of canal length measurements with film-based and digital systems. At the end of the day, are digital images more diagnostic than film-based images? Most studies suggest that DDR and film are not significantly different in their ability to record dental pathosis.19
Extraoral and panoramic digital imaging solutions are now available with enhancement technologies and more robust computer systems, which make them “relatively cost effective.”20 Future improvements may increase detection of systemic disease and improve ease of use.21 They include smaller sensors and automated image tools, digital subtraction, and caries diagnosis,22 as well as periapical and periodontal pathosis, three-dimensional viewing of teeth and facial structures, and bone analysis.
In a recent article by Haiter-Neto et al,23 100 extracted permanent human teeth were radiographically examined and compared using older and newer PSP and CMOS-APS systems. All of the teeth were without cavitations but exhibited a chalky white or brown area of discoloration. The exposure times at 65 kV peak and 10 mA with rectangular collimation for molars were 0.34 seconds for molars with the PSP and 0.26 seconds with the CMOS-APS systems. The images were then viewed by eight experienced observers with a software program that could adjust for contrast, brightness, gamma-curve, and magnification. The histologic evaluation of the teeth was then performed by two observers using a light microscope after embedding the teeth in acrylic, and serially dividing the teeth in 700-µm thick sections. The study found that the overall accuracies of the older and newer versions of PSP and CMOS-APS systems were not statistically significant, but there were more false-positive results with the newer PSP plates. The authors go on to state that the differences between all four systems may not be clinically significant.
Additional consideration should be given to barrier techniques and disinfection of PSP and solid-state detectors. Although the holders for these systems have disposable and/or autoclavable parts, Hokett24 et al recommend double coverage of rigid sensors with either latex, plastic, or equivalent barriers. He went on to state that simple hygiene procedures were adequate to prevent cross-contamination with digital detectors.
COMPUTERS AND MONITORS
Almost any new computer will be capable of supporting the available CCD and CMOS-APS detectors and their attendant software. Of course, choosing a central processing unit (CPU) that has the highest reliability and smallest form factor will provide significant life-cycle benefits. An example is the Dell™ OptiPlex™ series (Dell Inc, Austin, TX), which provides corporate-level reliability, ease of management, and scalable manageability. If a network is planned, then a server-based system is critical to ensure reliability.
Today’s flat-panel monitors are relatively inexpensive and significantly more energy-efficient than the now obsolete cathode-ray tubes. Recommended monitors such as the Dell 2405 (Dell, Inc), are sufficient for interpreting “small matrix” radiography images and cost 10 times less than medical-grade monitors.25
The purchase and continuation of hardware and software maintenance contracts are another area where diligence is important. The practitioner should create a system for monitoring the status of the maintenance plans for key elements of the IT system, especially digital radiography. Establishing insurance coverage for computer hardware, software, and data also requires special attention. General office insurance policies often limit computer coverage; therefore, practitioners should discuss policies with a knowledgeable insurance agent. In addition, staff should keep a file of purchase records, including offsite digital photographs of all equipment, to document purchases for insurance purposes.
High-quality inkjet and dye sublimation printers are capable of producing excellent “paper” records of grayscale radiographic and color photographic images. However, the printed image should not be considered diagnostic. Both types of printers have unique characteristics, and the choice may depend on cost and speed. These printers are useful for correspondence with colleagues who are not equipped to receive electronic images, and for documentation to insurance carriers. Eventually, all insurance records will be digitally transmitted, but the industry has been slow to adopt this technology.
Digital radiography reduces radiation, improves workflow, and allows new tasks to be performed. The promise of future improvements, including computer-aided diagnosis, will continue to add value for practitioners and their patients, facilitating operative procedures such as endodontics, oral surgery, and implant placement.
References1. The National Council on Radiation Protection and Measurements Report No. 145. Available at: http://www.ncrponline.org/Publications/145press.html. Accessed April 14, 2008.
2. Miles DA. Digital radiography. Inside Dentistry. 2006;2(5):82.
3. Kump KS, Omernick J. Assessment of image quality among multiple output devices using digital x-ray radiographs (abstract #832). Paper presented at: the Radiological Society of North America (RSNA) 2000 Conference; November 26-December 1, 2000; Chicago, IL.
4. GE Medical Systems, Inc. X-ray education: Digital x-ray. Available at: www.gehealthcare.com/usen/xr/edu/products/introdigdettech.html. Accessed April 14, 2008.
5. Paurazas SB, Geist JR, Pink FE, et al. Comparison of diagnostic accuracy of digital imaging using CCD and CMOS-APS sensors with E-speed film in the detection of periapical bony lesions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2000;89(3):356-362.
6. Litwiller D. CCD vs. CMOS: Facts and fiction. 2001. Photonics Spectra. Available at: www.dalsa.com/shared/content/Photonics_Spectra_CCDvsCMOS_Litwiller.pdf. Accessed on: April 14, 2007.
7. Sanderink GC, Miles DA. Intraoral detectors. In: Miles DA, ed. Applications of digital imaging modalities for dentistry. Dent Clin North Am. 2000;44(2): 249-255.
8. Yoshiura K, Kawazu T, Chikui Tet al. Assessment of image quality in dental radiography, part 2: optimum exposure conditions for detection of small mass changes in 6 intraoral radiography systems. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1999;87(1): 123-129.
9. Bender IB, Seltzer S. Roentgenographic and direct observation of experimental lesions in bone: I. 1961. J Endod. 2003;29(11):702-706.
10. Nair MK, Ludlow JB, Tyndall DA, et al. Periodontitis detection efficacy of film and digital images. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1998;85(5):608-612.
11. Farman AG, Avant SL, Scarfe WC, et al. In vivo comparison of Visualix-2 and Ektaspeed Plus in the assessment of periradicular lesion dimensions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1998;85(2):203-209.
12. Haiter-Neto F, dos Anjos Pontual A, Frydenberg M, Wenzel A. A comparison of older and newer versions of intraoral digital radiography systems. J Am Dent Assoc. 2007;138(10):1353-1359.
13. Wenzel A. Digital radiography and caries diagnosis. Dentomaxillofac Radiol. 1998;27(1):3-11.
14. Wenzel A. Digital imaging for dental caries. Dent Clin North Am. 2000;44(2):319-338.
15. Svanaes DB, Moystad A, Risnes S, et al. Intraoral storage phosphor radiography for approximal caries detection and effect of image magnification: comparison with conventional radiography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1996;82(1):94-100.
16. Lamus F, Katz JO, Glaros AG. Evaluation of a digital measurement tool to estimate working length in endodontics. J Contemp Dent Pract. 2001;2(1):24-30.
17. Ong EY, Pitt Ford TR. Comparison of radiovisiography with radiographic film in root length determination. Int Endod J. 1995;28(1):25-29.
18. Mentes A, Gencoglu N. Canal length evaluation of curved canals by direct digital or conventional radiography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2002;93(1):88-91.
19. Wallace JA, Nair MK, Colaco MF, Kapa SF. A comparative evaluation of the diagnostic efficacy of film and digital sensors for detection of simulated periapical lesions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2001;92(1):93-97.
20. Farman AG, Farman TT. Extraoral and panoramic systems. Dent Clin North Am. 2000;44(2):257-272.
21. White SC, Yoon DC, Tetradis S. Digital radiography in dentistry: What it should do for you. J Calif Dent Assoc. 1999;27(12): 942-952.
22. Analoui M, Stookey GK. Direct digital radiography for caries detection and analysis. Monogr Oral Sci. 2000;17:1-19.
23. Haiter-Neto F, dos Anjos Pontual A, Frydenberg M, Wenzel A. A comparison of older and newer versions of intraoral digital radiography systems. J Am Dent Assoc. 2007;138(10): 1353-1359.
24. Hokett SD, Honey JR, Ruiz F, et al. Assessing the effectiveness of direct digital radiography barrier sheaths and finger cots. J Am Dent Assoc. 2000;131(4): 463-467.
25. Levin, M. EndoNet Consulting: Computer Peripherals. Available at: http://www.endonet.com/Tip25ComputersandPeripherals.html. Accessed on April 14, 2008.
|About the Authors|
|Martin D. Levin, DMD |
American Board of Endodontics
Adjunct Assistant Professor
College of Dental Medicine
Nova Southeastern University
Fort Lauderdale-Davie, Florida
Clinical Associate Professor
Department of Endodontics
Prosthodontics and Operative Dentistry
University of Maryland Dental School
Chevy Chase, Maryland
Shelly L. Lee, DDS, MS
American Association of Endodontists
Chevy Chase, Maryland
John M. Sturgeon, MD
American Board of Radiology
Virtual Radiologic Corporation