BJR
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
British Journal of Radiology (2004) 77, 465-471
© 2004 British Institute of Radiology
doi: 10.1259/bjr/18928981
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wade, C
Right arrow Articles by Brennan, P C
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wade, C
Right arrow Articles by Brennan, P C

Full Paper

Assessment of monitor conditions for the display of radiological diagnostic images and ambient lighting

C Wade, BSc and P C Brennan, PhD

University College Dublin, School of Diagnostic Imaging, Herbert Avenue, Dublin 4, Ireland


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Conclusion
 References
 
Diagnostic efficacy is related to viewing conditions. An increasing number of radiology departments are using workstations for reporting and it was the aim of this study to assess monitor performance and ambient lighting in areas allocated to soft-copy reporting and review. The study was performed in 4 Dublin hospitals and 20 monitors were examined. Using a Society of Motion Pictures and Television Engineers' (SMPTE) test pattern, maximum luminance, spatial uniformity of luminance, temporal luminance stability, brightness and contrast resolution (gamma), geometry and ambient lighting was assessed. The results demonstrated that although temporal luminance stability and spatial uniformity appeared to be at acceptable levels, maximum luminance and gamma value variations were noted, with maximum luminance and geometry values often not complying with published guidelines. Cleaning the monitor face had no impact. 90% of viewing areas had acceptable ambient lighting levels. The data presented demonstrate that monitors examined were not operating at optimal levels for all performance parameters and inclusion of regular assessments of monitors should be part of an imaging department's ongoing quality assurance programme.


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Conclusion
 References
 
Over the past 20 years, medical imaging has been undergoing a quiet revolution. In radiology, clinical images that were traditionally viewed on film are now being acquired digitally and presented to the radiologist for diagnosis using soft-copy display devices, most commonly the monochrome cathode ray tube (CRT) [1]. This type of display currently offers the best performance and is the most highly developed and reliable display in common use [2]. Newer technologies include flat panel and plasma displays, but the CRT currently has advantages over these alternatives, including emission of light in all directions, allowing viewing angles up to 180° efficient electron focusing, facilitating small (high resolution) electron beams and ability to produce luminance values of up to 10 000 cd m–2 [3].

Accurate diagnosis and review of patient images from CRT devices relies on optimal performance, and for radiologists and radiographers to be satisfied that the CRT devices are working optimally a number of parameters need to be regularly assessed:

Maximum luminance
The term maximum luminance as employed by users of CRTs, is not the maximum luminance that the display system can provide, but rather the highest luminance level at which the display system shall be operated and maintained over a specified work life [2]. This needs to be high enough to ensure sufficient stimulation of the retinal cones, thus enhancing visual acuity. Luminance changes in the CRT may occur due to processes at the thermionic cathode of the electron gun or to inappropriate adjustments of the contrast-brightness controls on the monitor [4]. No universal standard exists for a maximum level of luminance, although the American College of Radiology (ACR) has set a guideline level of 50 ft lamberts (171.3 cd m–2) [5].

Spatial and temporal uniformity of luminance
Luminance of the CRT needs to be as consistent as possible across the CRT, otherwise variations in image brightness may be due to the CRT rather than patient characteristics, however due to unavoidable variations in phosphor coating, some spatial changes in the luminance will exist. This is known as spatial non-uniformity of luminance [2]. It is also important that the luminance of the monitor demonstrates temporal uniformity so that it remains stable through time, particularly when an image is first displayed after a long period of displaying a blank screen [6].

Contrast and spatial resolution
Contrast resolution is determined by the gamma value of a monitor, which is defined by the ratio characterizing the relationship between brightness and grey scale value of the monitor. High contrast resolution enables differentiation of structures with similar but different attenuative properties [2]. Spatial resolution determines how much detail a monitor can display and discerning small details in an image requires high spatial resolution. Hard-copy viewing can achieve spatial resolution values of 15-line pairs per millimetre [7] and currently this is rarely achieved with soft copy viewing. It should be noted that inadequate resolution is not solely due to the display system, but also to image acquisition.

Geometry
Geometry refers to the positioning of pixels on the screen. If the pixels are not displayed in their proper position, the image will appear with "barrel" or "pincushion" distortion [6], which can be as a result of inadequate magnetic shimming from external influences, CRT electronic maladjustment or malfunction or static charge on the phosphor screen [4]. Geometry is measured as a ratio of height to width, and a square image displayed on a monitor should have a ratio of 1:1 [6].

As well as parameters directly related to the CRT, the ambient light of the viewing area has an important impact on image viewing. Excess brightness decreases the perceived image contrast and can result in specular or diffuse reflections. The World Health Organization (WHO) and the Commission of European Communities (CEC) have, respectively, set ambient light standards of <=100 lux at 30 cm and <=50 lux at 100 cm from the CRT [8, 9].

The current investigation follows on from a recently reported study, which demonstrated that brightness and uniformity of viewing boxes in a number of diagnostic imaging departments did not comply with published guidelines [10]. In the last number of years however, imaging departments have increasingly started to use soft-copy workstations and specifically CRTs for diagnosis. Our aim was to determine current viewing conditions in departments, performing soft-copy reporting, over a 4 month period, concentrating on the aforementioned parameters. Resultant values will be compared with available guidelines as summarized in Table 1Go.


View this table:
[in this window]
[in a new window]
  		Table 1. Published guidelines on image viewing parameters for workstation monitors

 

   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Conclusion
 References
 
A survey of local teaching hospitals established that six hospitals used workstation monitors for the reporting and review of general radiography images. Two of the six hospitals did not contain any test pattern on their workstations, so these could not be included in the study. All other hospitals participated and were randomly allocated the identification letters A, B, C and D. All monitors used for the review and diagnosis of general radiographic images within imaging departments in these hospitals were assessed and the number of monitors per hospital ranged from one to nine. Details and distribution of monitors are shown in Table 2Go.


View this table:
[in this window]
[in a new window]
  		Table 2. Details of monitor type in all hospitals. Specifications shown are those provided by the manufacturer

 
The Society of Motion Pictures and Television Engineers' (SMPTE) test pattern was used to assess monitor performance (Figure 1Go). This is well described in the literature and has been used previously [4, 6]. A Nuclear Associates (New York, NY) photometer, model 07-621, with an ambient light shield was used to measure luminance from the SMPTE image in candela per metres squared (cd m–2) and ambient light in lux. A principal physicist calibrated the photometer before any measurements were made [3].



View larger version (90K):
[in this window]
[in a new window]
  		Figure 1. Society of Motion Picture and Television Engineers' (SMPTE) test pattern. Areas where spatial uniformity of luminance was measured are identified by arrows.

 
A number of measurements were made based on the parameters outlined in the introduction using procedures described elsewhere [6] and outlined below. The luminance output of a monitor tends to vary when first turned on; therefore, all measurements were taken after 2 h of switching the monitor on [6].

Before any measurements were taken the monitor was checked for appropriate contrast and brightness settings by checking that the 5% and 95% squares on the test pattern was just distinguishable from the 0% and 100% squares, respectively [6]. This was the case for all monitors and no adjustments were necessary.

The following parameters were assessed:

All the above measurements were taken at the same time of the working day, and each monitor was assessed once a month for 4 months. The type of monitor was recorded along with any manufacturer specifications (Table 2Go). The existence of a quality assurance program was also recorded.

The efficacy of a cleaning program was examined. After the first measurements were taken, the monitor face was cleaned with a screen cleaner produced by Agfa Gaevert (Wien, Austria) and a soft cloth, following which all measurements were repeated.

Agreement between independent viewers of the spatial resolution patterns was assessed using the Kappa test [12].


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Conclusion
 References
 
For each month of the investigation, 47 measurements were obtained for each of the 20 monitors; this resulted in 3760 readings over the 4-month period of the study. A further 940 readings were taken following cleaning of the monitors.

Variations were seen across monitors for maximum luminance, spatial uniformity, gamma values and geometry with 80% of mean maximum luminance and 60% of mean geometry measurements failing to comply with guidelines set by the ACR [5] and American Association of Physicists in Medicine (AAPM) [13], respectively. All spatial uniformity measurements met the AAPM guidelines [13]. For all monitors, throughout the investigation, temporal luminance did not change and all of the spatial resolution line pairs were clearly visualized by all of the observers. The data gathered for each hospital are shown in Tables 3–6GoGoGoGo.


View this table:
[in this window]
[in a new window]
  		Table 3. Monthly mean values for each monitor over the 4-month period for Hospital A

 

View this table:
[in this window]
[in a new window]
  		Table 4. Monthly mean values for each monitor over the 4-month period for Hospital B

 

View this table:
[in this window]
[in a new window]
  		Table 5. Monthly mean values for each monitor over the 4-month period for Hospital C

 

View this table:
[in this window]
[in a new window]
  		Table 6. Monthly mean values for the Phillips PCR AC 500 monitor over the 4-month for Hospital D

 
90% of ambient light measurements were appropriately below the level of 100 lux specified by the WHO [8] however large variations between areas were demonstrated, with a variation factor greater than 30 between the highest (Hospital C) and lowest (Hospital A) ambient light levels (Tables 3–6GoGoGoGo).

No differences in monitor measurements taken before and following a cleaning regimen were demonstrated.

Kappa analysis demonstrated no interobserver differences.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Conclusion
 References
 
In modern imaging departments, radiologists and other clinicians are increasingly using workstations for making diagnoses and reviewing images [1]. Diagnostic efficacy relies heavily on the quality on the monitors on which the images are being viewed and problems with luminance or spatial resolution could affect interpretation [1]. To ensure the monitors are operating optimally, a large number of performance parameters must be checked at regular intervals and these are well described in the latest draft document of the AAPM—Assessment of display performance for medical images [13]. Whilst a variety of test images is available to assess individual parameters, it is important that the success of any ongoing quality assurance programme relies on tests being straightforward, not time-consuming, performed outside a laboratory environment and not involving expensive equipment. This is why the SMPTE image has been widely used by previous researchers and is commonly used in imaging departments [4, 6]. The image has certain limitations and these will be considered below, however, the adoption of this single pattern, which allows accurate measurement of a variety of important parameters with minimum time and expense should be encouraged within imaging departments.

The results from the current study demonstrated that although spatial uniformity of luminance was good for all monitors, mean maximum luminance values across the 4 month period varied substantially between monitors. Although it was reassuring to note that in all hospitals with workstations used for primary diagnosis (radiology) these workstations had higher luminance values than the other workstations, it was of concern that only three of the 15 radiology workstations had maximum values above the ACR guidance level of 50 ft lamberts (171.3 cd m–2) [5]. Variation was also shown between monitors produced by the same manufacturer. Data from the monitors of one manufacturer showed that whilst the VC 33 monitors in Hospital A demonstrated the highest level of luminance across all monitors, identical monitors in Hospital C failed to reach the ACR guidance level. Also in Hospital C although monitors 1 and 2 were both the same model from one manufacturer, one had a mean maximum luminance of over twice the other. This type of intramanufacturer variation was also demonstrated for another manufacturer in Hospital B, demonstrating that by having a workstation with similar or the same specifications from a particular manufacturer does not guarantee that they will operate at consistent luminance. The well-described limitation of colour monitors was also evident from maximum luminance measurements.

The results from the temporal luminance stability tests were encouraging, showing no change for all monitors, however when one considered luminance values across the 4 month period, some monitors were shown to exhibit minimum/maximum variation factors close to 2. These variations did not appear to be directly related to the level of mean maximum luminance since the monitor with the greatest monthly changes, a monochrome monitor in Hospital B had one of the higher mean maximum luminance levels (166.8 cd m–2), whilst the monitor with lowest luminance, in Hospital A showed little monthly fluctuation. These results emphasise the importance of carrying out luminance checks at regular intervals.

Geometry and gamma values varied between monitors with the former often failing to achieve the recommended ratio of 1:1 [6]. Gamma values ranged from 1 to 1.9, which would suggest that certain monitors demonstrate considerably higher contrast than others. It is difficult to judge whether this is clinically important or not, particularly since there are no published guidelines to refer to, however it is of some concern that Wang et al reported worse contrast thresholds for both high and low background brightness when gamma values were below 2 [14]. Only two viewing areas had ambient lighting levels above the guidance level of 100 lux proposed by the WHO [8], with one of these having a lighting level of 425.8 lux. This extremely high reading resulted from the workstation being beside a non-obscured window, emphasizing the importance of proper planning of workstation location. Neither of the two areas with excessive ambient lighting had dimmer switches, which could have helped lower the light levels. Overall the ambient lighting results, like the previous work of McCarthy [10] showed that levels were generally appropriate and the image information was not excessively affected by specular or diffuse reflection, which can reduce the contrast characteristics of the image.

The spatial resolution test patterns provided by the SMPTE image are inadequate. All of the patterns were clearly visualized by all of the observers on all of the monitors. To provide a more meaningful assessment of spatial resolution and the impact of poor luminance or geometry on resolution, a test pattern with greater line pairs per millimetre is required for both the high and low contrast resolution pattern. Alternatively, an additional resolution test pattern such as one of the TG-18 types described by the AAPM should be available on monitors [13]. The authors would therefore advise that the current spatial resolution test results should be treated with caution.

The cleaning programme offered no simple solution to the luminance problems described above and since the brightness and contrast settings of the monitor were considered acceptable according to the protocol of Parsons et al [6] (see Method and materials), it is obvious that regular quality assurance and servicing of the monitors is required. It was stated that the only quality assurance programme being regularly performed on all monitors was in Hospital C, and although this supposedly involved a 3 monthly examination using the SMPTE test pattern and changes were made as necessary, no quality assurance was performed throughout this 4 month study.

The current findings would suggest that more effective programmes with shorter time intervals are needed, the absence of which may be linked to the paucity of manufacturer guidance on the type and regularity of monitor quality assurance programmes. It is interesting to note that the monitoring programme described in this investigation, takes about 10 min per monitor and provides a range of useful information, although an additional test, as stated above, is required for spatial resolution assessment.

Wherever possible the current investigation relied on published guidelines for the evaluation of measured findings. Information from a range of bodies was sought since no single body had recommendations for all the parameters measured. It would be helpful for individuals involved in studies of this nature and those participating in departmental quality assurance programmes, if recommendations from a single body, using consistent methods of measurement could be available for all relevant parameters to facilitate appropriate judgements of gathered data.


   Conclusion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Conclusion
References
 
Data from the current investigation demonstrate that a number of monitors in the hospitals examined are not operating at optimal levels. Although it is currently unclear as to the precise impact suboptimal monitors will have an impact on the accuracy of diagnosis, the results would suggest that departments should consider regular quality assurance programmes to ensure that monitors are complying with published guidelines whenever available.

Received for publication June 6, 2003. Revision received October 2, 2003. Accepted for publication February 3, 2004.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Conclusion
 References
 

  1. Krupinski E, Roehrig H. Recent developments in evaluating soft-copy displays for medical images. Recent Res Devel Optical ENGG 2000;3:35–49.
  2. Roehrig H. The monochrome cathode ray tube display and its performance. In: Kim Y, Horri S, editors. Handbook of medical imaging. Bellingham, WA: SPIE Press, 2000:156–220.
  3. Wang J, Langer S. A brief review of human perception factors in digital displays for picture archiving and communication systems. J Digit Imaging 1997;10:158–68.
  4. Roehrig H, Willis CE, Damento MA. Characterisation of monochrome CRT display systems in the field. J Digit Imaging 1999;12:152–65.
  5. American College of Radiology. ACR Standards for Teleradiology. In: Standards 2000–2001. Reston VA: American College of Radiology, 2001:9–17.
  6. Parsons DM, Kim Y, Haynor DR. Quality control of cathode-ray tube monitors for medical imaging using a simple photometer. J Digit Imaging 1995;1:10–20.
  7. Ball J, Price T. Chesney's radiographic imaging (6th edn). Oxford: Blackwell Science, 1995:316–7.
  8. WHO. Quality assurance in diagnostic radiology. Geneva, Switzerland: World Health Organization, 1982.
  9. Criteria for acceptability of radiological (including radiotherapy) and nuclear installations. Radiation Protection No. 91. Luxembourg: CEC, 1997.
  10. McCarthy E, Brennan PC. Viewing conditions for diagnostic images in three major Dublin hospitals: a comparison with WHO and CEC recommendations. Br J Radiol 2003;76:1–4.[Free Full Text]
  11. National Electrical Manufacturers Association. Standards for the performance of radionuclide imaging devices. Publication No. NU-1-1980. Washington DC: NEMA, 1980.
  12. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74.[CrossRef][Medline]
  13. American Association of Physics in Medicine. AAPM, Assessment of display performance of medical imaging systems, Report of the AAPM Task Group No. 18, 2002.
  14. Wang J, Anderson J, Lane T, Stetson C, Moore J. Contrast-detail characteristic evaluations of several display devices. J Digit Imaging 2000;13:162–7.[Medline]



This article has been cited by other articles:


Home page
 Br. J. Radiol.Home page
N Buls, W Shabana, P Verbeek, P Pevenage, and J De Mey
Influence of display quality on radiologists' performance in the detection of lung nodules on radiographs
Br. J. Radiol., September 1, 2007; 80(957): 738 - 743.
[Abstract] [Full Text] [PDF]

Home page
 Br. J. Radiol.Home page
D P Thompson, C J Koller, and J P Eatough
Practical assessment of the display performance of radiology workstations
Br. J. Radiol., April 1, 2007; 80(952): 256 - 260.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
P. C. Brennan, M. McEntee, M. Evanoff, P. Phillips, W. T. O'Connor, and D. J. Manning
Ambient Lighting: Effect of Illumination on Soft-Copy Viewing of Radiographs of the Wrist
Am. J. Roentgenol., February 1, 2007; 188(2): W177 - W180.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wade, C
Right arrow Articles by Brennan, P C
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wade, C
Right arrow Articles by Brennan, P C

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
BJR DMFR IMAGING  ALL BIR JOURNALS