Kenneth A. Fetterly, MS

The goals of a monitor quality control (QC) program are to ensure consistent display performance (thus allowing optimal image display), to identify and resolve problems before they become clinically relevant, and to make monitor use and support as efficient as possible. In addition to the monitor itself, the electronic image display system consists of a video card and cable, gray-scalecalibration software, and clinical image display software. Environmental factors, such as ambient light and stray radio-frequency signals, also contribute to image display performance.

Nicholas J. Hangiandreou, PhD

Major image-quality considerations for monitor QC include contrast, resolution, artifacts, and noise. It is especially important to the quality of diagnostic image interpretation (and, therefore, to patient care) to ensure that monitors have appropriate contrast throughout the full video range and that all monitors attached to a single workstation have the same grayscale response. The luminance range of the monitors should also meet the vendor's specifications, the recommendations of professional radiology societies, and the predetermined policies of the health care enterprise.

PROGRAM COMPONENTS

Steve G. Langer, PhD

A basic monitor QC program requires five primary tools: a predefined set of QC procedures, a set of appropriate test images, control limits, a database for tracking program activities and assets, and a photometer. Written QC procedures must be thorough, concise, and accurate.¹

An organized QC program has administrative advantages over crisis-based maintenance, but regular attention also improves the performance of monitors (Figure 1). For example, the ability to prevent the steady loss of cathode ray tube (CRT) luminance over time is of benefit. Several other common display problems can also be detected and corrected with an effective QC program. These include degraded sharpness, poor contrast, horizontal tearing, phosphor burn-in, and other problems typical of electronic displays.

Figure 1. Luminance degradation over time in monitors with and without regular maintenance.

Prior to performing QC procedures, the technician must first clean the monitor face and verify that viewing conditions are normal. In addition, he or she should verify that the screen saver is active, since this is an important factor in conserving monitor electronics and in preventing CRT phosphor burn-in problems. Note that implementation of a screen saver in conjunction with a flat panel liquid crystal display (LCD) would not be expected to lengthen its life-cycle because the back light will remain fully activated.

Our QC image series contains window patterns of 0%, 50%, and 100% luminance; a Society of Motion Picture and Television Engineers (SMPTE) pattern (Figure 2, page 50); a 100% uniform field pattern; and standard clinical images.  Objective measurements from the 0% and 100% video level images are obtained and the brightness and contrast of the monitors are adjusted accordingly to ensure an appropriate grayscale luminance range.  After adjustment, the luminance of a 50% video level is measured to verify that the overall contrast of the monitors is appropriate.  Subjective observations of whether the 5% and 95% SMPTE patches are visible, the image geometry is square, the images are spatially stable, and image flickering is minimal are also verified and recorded. Further subjective observations include evaluations of text sharpness, bar-pattern contrast, tearing at vertical edges, phosphor burn-in, and clinical image display.

Figure 2. A society of Motion Picture and Television Engineers test pattern used to monitor quality control.

All measurements have associated control limits that are based on clinical use, vendor specifications, and human perception. Subjective evaluations are based largely on experience.  We implemented a three point scale of like-new, adequate, and inadequate for the subjective tests.

Several more advanced monitor test patterns can also be used to assess display performance. The Briggs Pattern² (Figure 3a, page 52) or a Contrast-Detail Pattern can be used as a combined measure of monitor sharpness, noise, and contrast; the Cx Pattern³ (Figure 3b, page 52) evaluates sharpness alone. Recently, Task Group 18 of the American Association of Physicists in Medicine has made available recommended monitor evaluation procedures, including appropriate test patterns⁴ (Figure 3d, page 52). It is recommended that new monitor quality control programs be based on the work of this group. More advanced monitor characterization can be performed using a CCD camera.^5-7 Resolution measurements may be achievable with high-end consumer grade CCD cameras; however, measurements of monitor noise will require a scientific grade CCD camera.

Figure 3. A sample of monitor test patterns used to gauge various quality benchmarks includes (a) Briggs Pattern, used to assess sharpness, noise, and contrast; (b) Cx Pattern, used to assess sharpness; (c) Contrast-Detail Pattern, used to assess sharpness, noise, and contrast; and (d) TG18-QC Pattern(s) (American Association of Physicists in Medicine TG-18), used to assess geometric distortion, sharpness, noise, contrast, and contrast resolution.

Acceptance testing upon the arrival of a new monitor is an important part of a monitor QC program. This process not only ensures that the new monitor is working as well as it should, but also allows the technician to record baseline data for later comparison during periodic QC visits. At our institution, the objective measurements performed during acceptance testing include the standard QC measurements as well as static artifact evaluation of the glass and phosphor,⁸ and measurement of luminance overhead.  Also, it is appropriate to verify that the vendor's specifications of monitor performance are met.

The ability to accurately record and recall QC measurements is also important.  This will allow comparison of a monitor's QC data with previous measurements and comparison to the performance of other monitors. A database of QC results has been found to be very valuable for these purposes. Implementation of the database on a laptop computer provides a convenient means to enter data and perform real-time comparison to control limits in the field.

PROGRAM EXPERIENCE

Our formal QC program for monitors has been in place for 5 years. Covered by the program are 10 clinical quality assurance workstations with 10 monitors, 10 grayscale diagnostic workstations with 30 monitors, and 50 grayscale review workstations with 90 monitors.

Acceptance testing is performed prior to the installation of a new CRT monitor and QC is performed after a week, a month, and every 3 months thereafter.  These QC intervals have been found to be appropriate for our current CRT display technology; however, customization of these intervals dependent upon display type is appropriate. A monitor QC program can be planned using a staff-time estimate of 20 minutes per monitor per visit. This allows time for QC procedures (including luminance adjustments), travel between workstations, and data recording. For monitors on a quarterly QC protocol, this requires 80 minutes per year. We have found that CRT displays require additional maintenance after they are initially installed and that the life-cycle of the displays is approximately 2 years. Therefore, an overall estimate of approximately 3 hours of staff time per monitor per year is realistic for planning purposes. Given the purchase cost of electronic displays, it is appropriate to actively maintain these devices to ensure optimal performance and maximum life.  The costs associated with active support have been estimated to be a relatively small fraction (5-10%) of the overall cost of ownership of CRT displays.

NEW TECHNOLOGIES

The introduction of new display technologies will likely result in changes to QC procedures for electronic displays. As flat-panel liquid crystal displays continue to replace CRTs, the QC program and its resources will need to be adjusted accordingly. Improved luminance stability is expected from LCDs and image sharpness is expected to remain static. Often, automatic control of the luminance response is built into the display. It can be reasonably anticipated that LCDs will not require as much support effort as CRT displays. The decreased effort required to support LCDs will likely not offset the higher purchase price of these displays.  However, LCDs may prove to be cost-effective if their life-cycles are appropriately long.

CONCLUSION

For a monitor QC program to be effective, it must take into account the capabilities and limitations of the entire display chain. This requires that technicians be familiar with all of an enterprise's monitors, video cards, calibration software, and image-display software, as well as with external equipment that might create image display problems. Working relationships that foster the best possible support for monitor users must be established along with appropriate QC procedures. It has been our experience that the superior performance and increased life-cycle of monitors that are actively supported as part of a QC program justify the effort and resources expended.

Kenneth A. Fetterly, MS, Nicholas J. Hangiandreou, PhD, and Steve G. Langer, PhD, are medical physicists at Mayo Clinic, Rochester, Minn. This article has been adapted from Developing an Enterprise-Wide Monitor QC Program, which Mr Fetterly presented at the annual meeting of the Society for Computer Applications in Radiology in Salt Lake City,  May 2002.

Further Reading

National Institute of Display Technologies. Technology center. Available at: http://www.nidl.org/tech_mstr.htm. Accessed March 5, 2003.

Video Electronics Standards Association. Standards. Available at: http://www.vesa.org/standards.htm. Accessed March 5, 2003.

References:

1. Groth DA, Bernatz SN, Fetterly KA, Hangiandreou NJ. Cathode ray tube quality control and acceptance testing program: initial results for clinical PACS displays. RadioGraphics. 2001;21:719-732.
2. Briggs SJ. Softcopy display of electro-optical imagery. Proc SPIE. 1987;762.
3. Kohm KS, Cameron AW, Van Metter RL. Visual CRT sharpness estimation using a fiducial marker set. Proc SPIE. 2001;4319:286-297.
4. American Association of Physicists in Medicine Task Group 18. Assessment of display performance for medical imaging systems (draft). Available at: http://deckard.mc.duke.edu/~samei/tg18. Accessed March 5, 2003.
5. Roehrig H. The monochrome cathode ray tube display and its performance. In: Kim Y, Horii SC, eds. Handbook of Medical Imaging. Vol 3. Bellingham, Wash: SPIE Press; 2000:155-220.
6. Kruger RL, Fetterly KA, Hangiandreou NJ, Van Metter RL Determining the sharpness of electronic displays: an evaluation of three methods. J Digital Imaging. 2001;2:83-91.
7. Samei E, Flynn MJ. A method for in-field evaluation of the modulation transfer function of electronic display devices. Proc SPIE. 2001;4319:596-607.
8. Copeland JF, Helhus CS, Horton R, Venkatakrishnan V, Zamenhof RG, Mendel JB. Practical quality control standards for digital display monitors. Proc SPIE. 2000;3976:315-323.
9. Flynn MJ, Badano A. Image quality degradation by light scattering in display devices. J Digital Imaging. 1999; 2:50-59.