Quality Control Notes

QUALITY CONTROL

• Quality Assurance:
  • Managerial philosophy encompassing all aspects of patient care, image production, and image interpretation.
• Quality Control:
  • Refers to calibration and monitoring of equipment.
  • Staff radiographers can perform several tests.
  • Some tests require special equipment and should be done by a medical physicist, especially in the digital age.

COMPONENTS OF A QUALITY CONTROL PROGRAM

• Acceptance testing at the time of purchase.
• Establishment of baseline performance parameters.
• Diagnosis of correctable deviations in performance.
• Documentation of actual corrections made.
• Guidelines for testing and suggested parameters are published by AAPM and related organizations.

EQUIPMENT TESTS: FREQUENCY AND PARAMETERS

• Exposure Timer Accuracy:
  • Frequency: Annually
  • Acceptable Range: \pm 5%
• mA Linearity:
  • Frequency: Annually
  • Acceptable Range: \pm 10% adjacent stations
• Exposure Reproducibility:
  • Frequency: Annually
  • Acceptable Range: \pm 5%
• Kilovoltage Peak Calibration:
  • Frequency: Annually
  • Acceptable Range: \pm 5%
• Half-Value Layer / Filtration:
  • Frequency: Annually
  • Acceptable Range: Exceed required minimum
• Collimation:
  • Frequency: Semi-annually
  • Acceptable Range: \pm 2% of the SID (Source to Image Distance)
• Vertical Beam Alignment:
  • Frequency: Semi-annually
  • Acceptable Range: \pm 2 degrees of vertical
• Focal Spot Size / Condition:
  • Frequency: Annually
  • Acceptable Range: \pm 50% of nominal size
• Image Receptor System:
  • Frequency: Semi-annually
  • Acceptable Range: Zero visible defects
• Automatic Exposure Control:
  • Frequency: Annually
  • Acceptable Range: \pm 10%

RADIOGRAPHIC UNITS

• Exposure Timer:

  • Checked with a photodiode device or ion chamber.
  • Check after major repair or change.
  • Guideline: \pm 5%

• mA Linearity:

  • Accuracy of a particular mA station relative to the two adjacent mA stations.
  • Must be within 10% of each other.
  • Least reliable technique factor and most difficult to calibrate.
  • Guideline: 10% rule applies only to adjacent stations.

Calculating Percentage of mA Linearity:

• Subtract the difference in radiation output between two adjacent mA stations.
• Divide by the lesser of the two outputs.
• Multiply by 100 to get the percentage.
• Example:
  • 100 mA station: 6.4 mGy
  • 200 mA station: 11.2 mGy
  • 6. 4/100 = 0.064 mGy/mAs = 64 µGy/mAs
  • 7. 2/200 = 0.056 mGy/mAs = 56µGy/mAs
  • \frac{64 \mu Gy/mAs - 56 \mu Gy/mAs}{56 \mu Gy/mAs} = 0.143
  • 0.143 \times 100 = 14.3%
  • The 200 mA station is 14% out of calibration, outside the 10% limit.

Exposure Reproducibility:

• Ability to repeat the same technique settings and obtain the same results.
• Ten exposures are made using identical technique settings.
• The average exposure is calculated.
• Each individual exposure is checked against the average.
• Guideline: \pm 5% from average.

Half-Value Layer (HVL):

• Thickness of absorber material (usually aluminum) required to reduce the intensity of the x-ray beam to one-half its original.
• The only true measure of the actual penetration capability of the x-ray beam.
• Beam penetration may be raised by:
  • Adding filtration, which raises the minimum energies.
  • Increasing kVp, which raises the maximum energies.
• Any ion chamber can be used with a series of aluminum slabs of varying thicknesses, to determine the HVL

HALF-VALUE LAYER

• The most important quality of the x -ray beam is its ability to penetrate through the human body to produce an adequate image
• Minimum required HVLs are provided in tables published by regulatory agencies and scientific groups, particularly the NCRP (National Council of Radiation Protection and Measurement), Report #102
• The HVL must be specified for each type of generator and for each level of kVp

HALF-VALUE LAYER

• High HVL indicates that the x-ray beam is highly penetrating
• If the measured HVL falls below the required amount, it can indicate that either:
  • The calibration of the kVp control is off
  • There is insufficient filtration in the x-ray beam
• The kVp can be ruled out as the cause by using electronic calibration tools
• Once the kVp is calibrated, the only remaining cause of insufficient HVL can be insufficient filtration
• Insufficient HVL overrules the general minimum filtration rule - If kVp is accurate and the HVL is insufficient, filtration must be added until the HVL is adequate, even if it is more than 2.5 mm Al

RADIOGRAPHIC UNITS

• kVp Calibration:
  • For digital equipment, done by physicist using:
    • Ion chamber or photodiode with filtration
    • Voltage diode instrument or oscilloscope
  • Guideline: \pm 5%
• Collimator:
  • Edges of actual x-ray field can be off from the edges of the projected light field (1 cm), due to:
    • Belt or gear slippage
    • Mirror crooked
  • Guideline: \pm 2% of SID

COLLIMATOR AND ALIGNMENT CHECK

• Device: Plexiglass cylinder with metal bead for light field centering.
• Process: Collimate field edges to outlined rectangle.
• Evaluation: Light field alignment to x-ray exposure field; central ray verticality (guideline: \pm 2 degrees).

“HOME-MADE” COLLIMATOR CHECK

• Field center and edges can also be checked by using simple metal objects such as paper clips and markers
• The center of the x-ray field can be located by marking an “X” across the diagonal corners of the exposed field
• The same can be done with the corners of the metal objects to locate the center of the light field

RADIOGRAPHIC UNITS

• Distance:
  • Measure from the red “+” sign or other mark on the anode end of the x-ray tube housing, which designates the exact location of the focal spot
  • Guideline: \pm 2% of SID
• Focal Spot Blooming, Size and Condition:
  • Regulations allow actual focal spots under 0.8 mm to be 50% larger, and those over 0.8 mm to be 40% larger, than the nominal (advertised) size
  • Actual focal spot size is larger at higher mA stations, due to blooming of the electron cloud from mutual repulsion of more electrons
  • Sudden deviations in FS size indicate damage to the anode

FOCAL SPOT CHECK

• The “Slit” Camera:
  • A lead template with finely cut slits of decreasing width
  • Lines are placed perpendicular to the FS dimension being measured
  • Exposure is scanned to find the blur point in each direction
  • A table converts the blurred grouping into LP/mm

FOCAL SPOT CHECK

• LP/mm = Line Pairs per Millimeter
  • A measure of sharpness or resolution
  • The greater the resolution, the smaller the focal spot
• Once a baseline measurement is recorded, any sudden loss of resolution indicates damage to the focal spot

RADIOGRAPHIC UNITS

• Automatic Exposure Control (AEC):
  • Reproducibility should fall within \pm 10%
  • AECs should also be linear within 20% between different rooms
  • Back-up timer can be tested by placing leaded rubber sheets over all detectors and seeing if the exposure terminates automatically at the back-up time or at 600 mAs

FLUOROSCOPIC UNITS

• Since patient exposure levels can be very high, QC tests must be performed regularly and accurately by a medical physicist, to include “spot-film” devices exposure levels and collimation, and AEC function
• A medical physicist should perform calibration checks after any major change or repair of the tube, generator or console
• Simple observational tests for distortion can be performed by the radiographer, including checks for pincushion and S-distortion, vignetting and veiling glare, described in Chapter 38

FLUOROSCOPIC UNITS

• Annual checks should include:
  • Minimum source-to-table distance of: 38 cm (15 inches) for stationary units; 30 cm (12 inches) for mobile units
  • 2 mm lead equivalent shielding in fluoro tower
  • Proper filtration
  • Proper automatic collimation
  • Presence of bucky slot cover
  • Presence of fluoro curtain
  • Function of cumulative timer (5-minute)
  • Output within guidelines (<10 R/min)
• In some states, the radiographer must record the amount of fluoroscopic beam-on time for each procedure, and in some, the number of overhead and spot exposures taken

MONITORING OF DIGITAL ACQUISITION SYSTEMS

• Radiographers can monitor sudden changes in field uniformity, erasure thoroughness, intrinsic noise and spatial resolution by visual checks
• Field Uniformity:
  • Digital detectors are all inherently non-uniform
  • Take a flat-field exposure with no object in the beam at a moderate technique
  • Open collimation to cover the entire field, and use a long 180 cm SID
  • Scan the resulting image for lighter areas or white specks

Erasure Thoroughness and “Ghosting”:

• Ghost images can indicate:
  • Bulbs in the reader are burned out
  • Loss of lamp intensity in the reader
  • Too short exposure to lamps in the reader
• They can also occur from extreme overexposure during the previous procedure (when residual electric charge is trapped)
• Expose the imaging plate with a step-wedge or other homogeneous object in the beam, (and process if CR)
• Immediately re-expose the same plate without the object in the beam and with 1 inch (2.5 cm) collimation in from each edge
• Scan the resulting image for any ghost image of the object

Intrinsic (Dark) Noise:

• Erase and then immediately process a single plate without exposing it to an x-ray beam
• Scan the resulting image for any unusual amounts of mottle or noise compared to a baseline image

Spatial Resolution:

• A flat wire mesh can be exposed, and the image examined for any distortion or blur (as at the bottom in the mesh)

Spatial Resolution:

• A lead foil “bar” test pattern can be used to obtain spatial resolution in LP/mm

MONITORING OF ELECTRONIC IMAGE DISPLAY SYSTEMS

• The display monitor is the weakest link in the imaging chain
• Consistency between all monitors within a workstation is essential
• All should have the same luminance, be set at the same contrast, and be cleaned monthly
• Guidelines and tests are available from several groups, including:
  • AAPM (American Association of Physicists in Medicine)
  • SMPTE (Society of Motion Picture and Television Engineers)
  • ACR (American College of Radiology)

MONITORING OF IMAGE DISPLAY SYSTEMS

• Various test patterns are available from Task Group 18 of the AAPM
• This universal, generic test pattern is available from SMPTE (The Society of Motion Picture and Television Engineers) -It includes checks for most image qualities all in one template

MONITORING OF IMAGE DISPLAY SYSTEMS

• Class 1 display monitors:
  • Are those used for diagnosis
  • These are subject to much more stringent guidelines than Class 2 monitors
• Class 2 display monitors: may be used for display and a limited amount of image manipulation, but not for documented diagnosis

LUMINANCE

• The rate of light emitted from an LCD, LED, or CRT monitor
• The unit for luminous flux (flow) as perceived by the human eye is the lumen
  • 1 lumen = 0.0015 Watts of power
• For the unit lumen to be meaningful, we must define the area of space that this amount of power is emitted across, or its concentration in space – This is the steradian

Steradian

• A steradian is the cone swept out by a 3D (solid) angle such that the area of its base is equal to the square of its radius from the light source (r^2)
• There are roughly 6 steradians in a hemisphere (12.57 in a sphere)

THE CANDELA (CD)

• The total rate of light emitted by a typical candle in all directions
• The lumen refers to the light flux a typical candle emits within just one steradian
• The candela describes the power of the light source, whereas the lumen describes the power of the light flow (flux) traveling through space

Candela and Lumen Relationship

• One candela generates one lumen per steradian
• 1 Cd = 1 \frac{Lm}{sr}

EXAMPLES OF LUMINANCE

• Maximum for an LCD = 800 Lm
• Maximum for a CRT = 600 Lm
• Maximum for a conventional viewbox = 3000 Lm
• Typical range preferred by radiologists = 500-600 Lm on electronic display monitors

MEASURING LUMINANCE

• LCDs need only be checked once each year
• The photometer is a device used to measure the light intensity from a display monitor

MEASURING LUMINANCE

• Photometer read-outs are usually in units of lux (to be defined shortly), lumens or candela / m^2

ILLUMINANCE

• Refers to the rate of light striking a surface, or how well objects in our field of view are illuminated by a light source
• e.g., the luminance of a desk lamp illuminates a paper on the desk in front of it

Illuminance Units and Typical Levels

• The unit for illuminance is the lux, defined as 1 lumen per square meter of surface area
• 1 lux = 1 \frac{Lm}{m^2}
• Typical illuminance levels:
  • Direct sunlight: 10^5 lux
  • Overcast sunlight: 10^3 lux
  • Full moonlight: 10 lux
  • Typical office lighting: 75-100 lux
  • Radiologic reading room lighting: 2-25 lux
• A radiologic reading room should have less than ¼ normal office lighting: 25 lux or less

LUMINANCE AND CONTRAST TESTS

• Maximum luminance: Check with photometer
• Compare with baseline over time: Check LCD yearly
• Luminance response: A monitor’s ability to accurately display different levels of brightness from a test pattern
• Essentially identical to a contrast test

Luminance Response Test

• SMPTE test pattern for luminance response (contrast): Adjacent squares in outer ring should all be distinguishable, along with inner squares of 0/5% and 95/100% .
• 50% squares should match.

DICOM GSDF (GRAYSCALE STANDARD DISPLAY FUNCTION)

• The DICOM GSDF is based on human perception, with increments of brightness called JND’s (Just noticeable differences)
• AAPM recommends that luminance response should fall within 10% of the GSDF standard

DICOM GSDF Standard

• Adjacent squares that are not distinguishable indicate insufficient contrast
• Adjacent squares on the SMPTE test pattern should measure 10% difference between each other on a photometer, within 10% accuracy

LUMINANCE AND CONTRAST TESTS

• Luminance Ratio: Maximum luminance divided by minimum luminance
• \frac{L{MAX}}{L{MIN}}
• An LCD is not capable of producing a true black level while it is energized
• This can be seen in a completely darkened room

LUMINANCE AND CONTRAST TESTS

• LCDs have a relatively poor luminance ratio
• On the SMPTE universal test pattern, black-on- white and white-on-black bars are used to measure LR

LUMINANCE AND CONTRAST TESTS

• Luminance Uniformity: Consistency of a single brightness level displayed across the area of the display screen
• AAPM recommends that luminance uniformity not deviate more than 30% from the average
• Luminance Uniformity is checked at 5 locations across the screen

AMBIENT LIGHTING AND REFLECTANCE

• Diffuse reflectance is the ratio of the brightness of the general ambient light in the room that is being reflected off the surface of the display screen to the output brightness of the monitor itself
• Specular reflectance refers to the reflection of a specific, localized light source behind the viewer of the display monitor, and is defined as the ratio of the reflected brightness to the brightness of that light source

AMBIENT LIGHTING AND REFLECTANCE

• In a diagnostic reading room, ambient lighting must be dimmed to a point where both diffuse reflectance and specular reflectance are below any noticeable level
• LCDs generally have low reflectance

ELECTRONIC DISPLAY RESOLUTION

• Both the ACR and the AAPM recommend a minimum resolution for electronically displayed images of 2.5 LP/mm
• On any test pattern for spatial resolution, series of high- contrast bars of diminishing width are used:
  • Horizontal bars measure vertical resolution
  • Vertical bars measure horizontal resolution

SMPTE Pattern for Resolution

• The universal SMPTE pattern has resolution bars in 5 locations:
  • Center, and 4 corners
• Each series consists of a set of high-contrast bars and a set of low-contrast bars

DEAD AND STUCK PIXELS (LCD)

• The LCD monitor can be visually inspected for faulty pixels and subpixels with the aid of a simple magnifying glass
• For most LCDs, one pixel is just the size of a 12-point font period or the dot of an “i”
• A defect smaller than this would be caused by the failure of a subpixel
• There are three rectangular subpixels in each pixel

DEAD AND STUCK PIXELS (LCD ONLY)

• A truly dead pixel appears as a white spot against a solid black background
• A stuck pixel appears as a black spot against a solid white background
• A stuck pixel is being continuously supplied with electricity

CRITERIA FOR DEAD AND STUCK PIXELS

• For a Class 1 LCD monitor, the AAPM recommends as “passing criteria:
  • 15 or fewer bad pixels across the entire screen
  • Three or fewer bad pixels within any 1-cm circle
  • Three or fewer bad adjacent pixels
• Less stringent guidelines apply for class 2 monitors

VIEWING ANGLE DEPENDENCE (LCD)

• More expensive LCD monitors employ pixel elements that are slightly angled in order to increase the effective viewing angle
• VAD can be measured using a photometer at a precisely maintained distance from the center of the monitor screen, as a percentage of the perpendicular luminance
• Rapid loss of luminance at increasing viewing angles is one of the major disadvantages of LCDs

REPEAT ANALYSIS IN THE DIGITAL WORLD

• An unfortunate aspect of digital systems is that since rejected images are usually erased and permanently lost, there is no way to track them and follow up with targeted “repeat” analysis
• Because the advent of digital imaging has dramatically reduced the rate of repeated exposures due to poor technical quality, we can state that most retakes are now due to positioning errors

REPEAT ANALYSIS IN THE DIGITAL WORLD

• A realistic target for diagnostic x-ray departments is to maintain an overall repeat rate of 3-5% due to technologist errors