In general, digital detectors exhibit the ability to separate the three major attributes of the x-ray image: acquisition, display, and archive. Wide exposure latitude and subsequent contrast enhancement allow under-? or over-exposure of the incident radiation, most often without the need to repeat the examination. The degree of contrast enhancement is ultimately limited by the signal-to-noise ratio (SNR), which is chiefly determined by the number of x-rays contributing to the formation of the image. From the physics perspective, the detective quantum efficiency (DQE) measures the detector information capture efficiency, which is dependent on the characteristics of the x-ray absorption, electronic noise, and structured noise. With a higher DQE, the radiation levels required to achieve a given SNR are reduced, as is patient dose. Increased thickness of the x-ray detector can improve the DQE in an “unstructured” phosphor such as barium fluorobromide (BaFBr) used in computed radiography (CR), or gadolinium oxysulfide (Gd2O2S) used in direct radiography detectors, but usually with reduced spatial resolution due to spread of the secondary light photons collected. For CR, new technology dual-side readout of the phosphor plate increases the amount of light photons captured during the laser-beam scan, and can improve DQE significantly without a loss of resolution. Structured phosphors such as cesium iodide (CsI) can achieve good detection efficiency and high spatial resolution with greater phosphor thickness simultaneously by channeling light inside individual crystal “light-pipes,” which reduces secondary light spread. Structured phosphors, however, are more fragile than unstructured phosphors and can be more readily damaged.

Direct acquisition photoconductors using an amorphous selenium photoconductor achieve high intrinsic resolution relatively independent of thickness because the electron/hole pairs generated by the absorbed x-rays are actively acquired under a high voltage across the photoconductor material, limiting the spread of the electrons to within the corresponding detector element. However, since absorption efficiency is strongly dependent on high atomic number elements, the relatively low atomic number of selenium requires a much larger thickness of material to achieve necessary x-ray absorption efficiency for high kVp examinations (eg, chest x-rays). Charge trapping, latent image retention, and lag effects can reduce the effectiveness of the direct detector capabilities.

Table I. General comparison: advantages/disadvantages of CR, DR, and CCD. Note: **** most desirable; * least desirable.

Besides the spatial resolution/dose issues, other digital system comparisons should include equipment costs, positioning flexibility, ease of use, and service/preventive maintenance among other considerations. Add-on or passive detectors such as CR provide the best flexibility and do not require the purchase of integrated x-ray systems. DR devices are generally more expensive, but typically can provide higher patient throughput and have excellent x-ray system interfaces that allow complete study descriptions (eg, mAs, kVp values) in the DICOM header and actually can set up the techniques for the technologist based on the examination type through the same modality worklist functionality. Purchase decisions must be made on a case-by-case basis, as all digital radiography detectors currently available have strengths offset by weaknesses, some of which are listed in Table I.

Life-cycle costs

CR requires continual maintenance and evaluation of imaging plates and cassettes; negotiation of an appropriate warranty is important for expected lifetime. Periodic cleaning of the imaging plates is also necessary, the frequency determined by the environment in which the system is run. For instance, CR in the emergency department requires a high cleaning frequency of imaging plates, while CR digital mammography has an extremely low incidence of dust and therefore requires minimal cleaning. Over time, replacement of CR cassettes and imaging plates is necessary due to normal wear and tear. DR system life-cycle costs are still not very well known, as these systems have been clinically operational for only a few years. A detector malfunction could require replacement of a thin-film transistor (TFT) panel at an extremely high cost. In any case, the total cost should include the purchase price, maintenance, replacement of equipment (eg, CR plates and cassettes), operational overhead, patient throughput, and other intangible costs such as time required to process images, do periodic quality control, and do require user maintenance.

What happens when a digital system is inoperable? Contingency plans must be considered for essential and critical imaging services, and redundancy must be built into the implementation plan. One scenario might include the availability of CR in conjunction with DR to maintain the use of a dedicated x-ray room in spite of the loss of a digital detector, and multiple CR readers if an unexpected breakdown of a CR reader occurs.

Acceptance Testing and QC

Any imaging system is only as good as its weakest link. The maintenance of optimal quality requires a robust quality control program, which starts with the acceptance testing of these systems to determine and verify the “system” capabilities and to benchmark the performance, in terms of spatial resolution, SNR, image uniformity, and other system-specific tests. “Flat-field” preprocessing is extremely crucial for correcting detector imperfections and detector background variations for both CR (Figure 1) and DR (Figure 2) detectors. Digital detectors have noticeable flaws that, if left uncorrected, will result in unusable or highly compromised images. All digital detectors employ a correction/calibration scheme, and some require more frequent calibration than others. This can be as often as once per day or as little as once per year, depending on the type of detector and manufacturer, and should be a question asked during purchase negotiations. DR systems require a robust calibration effort to eliminate structured noise artifacts and nonfunctioning pixel elements, rows, and columns. As detector systems begin to age, variations in response require calibration updates to maintain optimal image quality and maintenance of high DQE. Persistence in performing quality control measurements and constant vigilance to ensure adequate imaging performance are a key to a successful implementation and long-term satisfaction with any digital system.

Figure 1. Shading variability of a digital mammography CR system shows background variations of a uniformly exposed imaging plate

Figure 1. Shading correction calibrations eliminate the scan direction variations and improve uniformity.

Regarding quality control issues, the manufacturers need to do a better job of providing digital radiographic phantoms with automated software to allow the objective performance testing of the equipment to ensure optimal system function. Digital detectors provide an excellent opportunity to use the computer system to quantitatively measure detector performance. Ideally, a reasonable evaluation could be performed using a “one-shot” QC phantom with specific acquisition techniques and analyzed using automatic phantom evaluation software. A “greenyellowred” system status for a technologist could indicate proper system functionality. In addition, a chronological plot of the results that can identify trends, indicate potential malfunctions, and implement corrective action prior to a major breakdown would be helpful. As potential purchasers of such equipment, it is important to express the need for such QC capabilities when writing specifications and/or negotiating purchase terms to initiate their development.

Figure 2. A flat-panel TFT array uncorrected image of two sets of step wedges shows dead detector elements, row and column defects, and variable subpanel detector sensitivity.

Figure 2. Flat-field correction techniques and contrast enhancement significantly improve the image quality.

The Final Word

With current technology offerings, CR is the most readily available and cost-effective technology for digital radiography and integration into a PACS; however, direct digital radiographic systems have distinct advantages with detection efficiency, throughput, and PACS integration. Digital solutions are likely best accomplished with a complementary mix of technologies, such as placing DR in high throughput areas such as dedicated chest rooms, and using CR for general radiography applications and portable imaging. Certainly, the starting point is to understand the benefits and capabilities of digital detectors, be aware of improvements in technology, keep track of system cost assessment, and base the decision of acquisition/deployment on an objective analysis of digital system attributes and required functionality.

Note: “The Digital Capture Question: Part I” was published in the June 2003 issue of Decisions in Axis Imaging News.

J. Anthony Seibert, PhD, is professor of radiology, University of California, Davis Medical Center, Sacramento.