Radiography Quality Control Tests: What Is Measured in DR and CR Systems?

An X-ray unit can degrade in image quality without ever failing: the tube voltage can drift from the set value, the AEC can terminate the exposure differently, or unnoticed artifacts can appear on the detector. The purpose of quality control (QC) tests is to catch these changes before they reach patient images. So QC is not only a physicist's concern: these tests sit behind the image a radiologist and technologist see every day, and for a physicist in radiotherapy or nuclear medicine it is valuable to see how diagnostic radiology is kept in check. In this article we take the radiography-specific tests one by one — what is measured, how, with what equipment, by whom, and how the result is judged — distinguishing DR (digital flat panel) from CR (storage phosphor) where relevant.

What is it?

In its simplest definition, quality control is the process of verifying, by measurement, that a radiography system's performance today is the same as it was yesterday. Technically, this means being able to show that every link of the imaging chain — generator, tube, collimator, AEC and digital detector — works within accepted limits. QC is the measurable, technical leg of a broader quality management system (QMS); the IAEA emphasizes that the medical physicist plays a central role in this system, especially in equipment performance.¹

Tests are of two kinds: acceptance testing before the device enters service (establishing baseline/reference values), and the constancy tests repeated regularly thereafter.¹

Why is it done?

Because an X-ray unit can drift over time without failing outright: calibration values can change, detector response can shift, the AEC can terminate differently. These drifts are invisible to the eye. QC has three aims: patient safety (avoiding unnecessary dose), diagnostic quality (consistent, reliable images) and regulatory compliance (mandatory performance limits). For example, minimum HVL is legally required in the US under 21 CFR 1020.30.²³

What happens when it drifts?

Every test has a "what if it's not done / drifts" counterpart:

• kVp deviation: If the actual kVp is lower than set, contrast may rise (the photoelectric effect's share increases) but, because fewer photons reach the detector, image noise also rises. If the actual kVp is higher, contrast falls and the dose balance shifts. Combined with the AEC this becomes systematic under- or over-exposure.
• Low HVL (insufficient filtration): the proportion of low-energy photons in the beam rises (the beam "softens"). These photons do not contribute to the image but raise skin dose.²
• AEC drift: inconsistent exposure with patient thickness; either noisy (low dose) or over-dosed images, and repeat exposures.
• Light–beam field mismatch: part of the anatomy falls outside the field (cutoff) or extra area is irradiated — both repeats and excess dose.
• DR uniformity/dead pixels/ghosting: fixed-pattern artifacts can mimic or hide a lesion.
• Incomplete CR erasure: a "ghost" of the previous image bleeds into the new one; risk of misinterpretation.¹

What is measured, and how?

Most generator and tube tests can be done with a non-invasive multimeter placed at the detector plane and a calibrated dosimeter.

1 · Tube voltage (kVp) accuracy. Using a non-invasive kVp meter at the detector plane, several kVp settings are exposed and the measured peak voltage is compared with the value set on the console. Deviation affects both contrast and dose.

2 · Radiation output: reproducibility and linearity. The reproducibility of the output across repeated exposures at the same technique is assessed via the coefficient of variation (COV);² linearity with mAs is checked by whether the measured output rises proportionally across different mAs values. 21 CFR 1020.30 sets limits for both.³

3 · Beam quality: half-value layer (HVL). Aluminum filters of increasing thickness are placed in front of the beam and the dose is recorded at each thickness; the filter thickness that halves the initial dose is defined as the HVL. The measured HVL must meet the minimum HVL requirement for that kVp (Bushberg Table 6-3; 21 CFR 1020.30).²³

4 · Automatic exposure control (AEC). The aim is to verify that even when phantom thickness or tube voltage changes, the exposure reaching the detector (and the associated image indicator) stays constant within acceptable limits.¹ Measurements are made at different thicknesses (e.g. PMMA / aluminum slabs) and different kVp settings.

5 · Light field – beam field congruence and collimation. The collimator's light field must coincide with the actual X-ray field; per Bushberg the misalignment, along the length or width of the field, must not exceed 2% of the SID.² Alignment of the beam field with the image receptor is also checked.

6 · Digital detector — DR. Two points stand out in DR:

• Exposure index (EI) and deviation index (DI): EI does not directly indicate patient dose; it indicates the exposure reaching the detector (the receptor signal). In standardized systems, a deviation index (DI) is used alongside EI: DI = 0 corresponds to the target exposure, positive DI means over-exposure and negative DI under-exposure.⁴ QC verifies that the system produces the correct EI and that the EI is repeatable/consistent.¹
• Uniformity, dead pixels and ghosting/lag: under a flat, uniform exposure the detector response should be uniform; dead/stuck pixels and residual signal from a prior exposure (lag/ghosting) are checked.¹

7 · Storage phosphor — CR. CR cassettes need extra tests:

• Plate erasure integrity: after each read, the previous image on the plate must be completely erased; incomplete erasure causes ghost images.¹
• Plate uniformity and sensitivity: uniform response across the whole plate under a flat exposure; consistency of sensitivity between plates.

8 · Spatial resolution and low contrast. Resolution is assessed with a line-pair phantom (or by measuring the MTF); the limiting resolution is the spatial frequency at which the modulation drops to about 5% of its initial value — i.e. where the MTF curve practically reaches zero.¹ Low contrast is examined with contrast-detail phantoms.

9 · Reject (repeat) analysis. The rate and causes of rejected/repeated images are tracked regularly; a rising repeat rate can be an early sign of a systematic problem in the AEC, technique factors or positioning.

What equipment is needed?

• Non-invasive kVp meter / multimeter — kVp, exposure time, output and linearity.
• Calibrated dosimeter + ion chamber — radiation output/dose measurement.
• Aluminum filter set — HVL measurement.
• Light–beam alignment test tool (alignment plate / mesh) and a ruler/markers.
• Line-pair resolution phantom and a contrast-detail phantom.
• A uniform-exposure setup and software for DR/CR uniformity and artifact analysis.
• Copper/aluminum sheets — phantom thickness / patient simulation (AEC tests).

Who does it, who checks?

In practice the division of labor is roughly:

• Daily/weekly routine (detector/EI checks, visual inspection, CR plate cleaning): radiologic technologists / technical staff.
• Comprehensive annual tests: in practice usually performed by the medical physicist together with an authorized service organization.
• Acceptance testing and drift investigation: the responsibility of the medical physicist; the IAEA emphasizes the physicist's central role in equipment performance.¹
• Oversight/sign-off: results are evaluated by the medical physicist; regulatory compliance falls under the competent authority.

How are results evaluated?

Each measurement is compared with the baseline value from acceptance testing and judged against predefined levels:

• Tolerance (acceptance) level: if within this range, the system is acceptable.
• Remedial level: if exceeded, investigation and correction are required.
• Suspension level: if exceeded, the device is withdrawn from clinical use until corrected.

In DR the same logic applies to EI/DI and the reject rate: a marked deviation from the target EI (DI) or a rising repeat rate triggers a protocol review. These levels and test frequencies are set by national regulations and protocols (IAEA, AAPM, EUR); the values here are examples — the facility protocol and competent authority take precedence.⁵

← All articles