Visual Test Patterns
SMPTE Diagnostic Imaging Test Pattern
In 1985, the Society of Motion Pictures and Television Engineers (SMPTE) published a Recommended Practice (RP-133) document entitled Specifications for Medical Diagnostic Imaging Test Pattern for Television Monitors and Hard-copy Cameras. The idea was to create a pattern that could be used for both quantitative and qualitative assessments of an image display system. The document describes the attributes and specifications of the test pattern, it does not provide the test pattern in digital, analog, or film form. The test pattern was intended to provide for initial acceptance testing as well as quality assurance purposes in a clinical setting.
The following is a general description and explanation of its design. The background is a uniform gray at 50% maximum gray level. This is used to detect any non-uniformity across the entire display. A grid pattern over the entire background is used to assess any geometric or spatial distortions. Contrast resolution patterns in the four corners and the center are to assess spatial resolution and sharpness of the image. White (100%) on black (0%) and black (0%) on white (100%) rectangles are used to assess sharp transistions of contrast. A gray scale is presented in 11 squares around the center to assess the entire dynamic range from 0% to 100% maximum gray level in 10% increments. Also present are two squares with 0% and 100% maximum gray levels with smaller squares located inside having 5% and 95% gray levels, respectively. This is used to assess contrast resolution at the upper and lower limits and to allow for adjustment of the contrast and brightness controls of the image display device. In order to adjust brightness/contrast of the display, both controls should be turned to their lowest setting, then the brightness should be increased until the pattern is just visible above the background. The contrast should be increased until the 95% square is visible in the 100% square.
AAPM TG18-QC Test Pattern
In 2005, the American Association of Physicists in Medicine (AAPM) published the Task Group 18 report entitled Assessment of display performance for medical imaging systems. This report suggested guidelines and criteria for acceptance testing and quality control of medical display devices. The report describes test patterns for both quantitative and qualitative assessment of display systems. Displays are categorized as either primary displays for diagnostic interpretation of medical images or secondary review displays for viewing by medical staff or specialists other than radiologists after an interpretative report is rendered. Necessary instrumentation, software, personnel, display placement, testing procedures, and viewing environment (including recommendation for ambient light levels) are defined. Assessment of display devices in a clinical setting using the multi-purpose TG18-QC test pattern to evaluate geometric distortion, artifacts, spatial resolution, dynamic range, contrast -detail resolution is specified.
The following is a general description and explanation of its design. The background is a uniform 109 gray level. This is used to detect any non-uniformity across the entire display. A grid pattern over the entire background is used to assess any geometric or spatial distortions. Sixteen 102 x 102 squares with gray levels varying from 8 to 248 are used for luminance response or dynamic range evaluation. Each square contains four small 10 x 10 corner boxes at ±4 gray levels difference from the background, +4 in upper left and lower right, -4 in lower left and upper right. Also present are two squares with 0 and 255 gray levels with smaller squares located inside having gray levels of 13 and 242, respectively. Line-pair patterns at the center and four corners at Nyquist and half-Nyquist frequencies are used for spatial resolution assessment. “QUALITY CONTROL” letters with various contrast at minimum, mid-point, and maximum gray levels are used to assess contrast-detail. Two vertical bars with continuous pixel value variation are used for evaluating bit depth and contouring artifacts. White and black bars are used for evaluating video signal artifacts.
