Artifacts can obscure anatomy, mimic disease, or prompt an avoidable repeat exposure. Digital acquisition also adds detector-calibration, sampling, exposure-field-recognition, and processing failure modes. Radiologic technologists should recognize likely patterns, preserve the original image and exposure information when investigating them, and follow the facility's quality-assurance (QA) pathway.
Useful source categories are patient or external objects, acquisition and geometry (motion, scatter, collimation, grid alignment, exposure), detector or reader hardware, and processing or display. Categories overlap, so appearance alone rarely proves a root cause.
First protect the patient: do not make an extra patient exposure merely to reproduce an artifact. Record the detector/cassette, reader, grid, protocol, orientation, and exposure indicator; inspect the equipment; then use an approved phantom or uniform-field QC test if authorized. A detector-fixed pattern supports a detector or plate cause, but reproducibility is not proof—processing, grid, and generator faults may also recur.
Computed radiography (CR) uses photostimulable-phosphor (PSP) imaging plates, usually housed in cassettes and scanned in a reader. Plate handling, cassette condition, transport through the reader, laser scanning, light collection, erasure, and processing each introduce potential artifact sources.
Physical damage to a photostimulable-phosphor (PSP) plate can produce fixed linear or irregular defects. Their displayed brightness depends on whether the damage removes phosphor, changes emitted light, or blocks readout, and on the selected grayscale polarity; therefore “radiolucent” is not a reliable universal description. Dirt, abrasive particles, reader transport, or handling can damage a plate.
Prevention: Handle, transport, inspect, and clean cassettes and plates at the manufacturer- and facility-specified intervals with approved materials. Remove a plate from clinical use when a fixed defect exceeds the local action criterion or could obscure diagnostic information; physical loss of phosphor is not corrected by erasure or processing.
Electrostatic discharge can create branching “lightning” patterns in cassette-based imaging. Dry conditions and handling can promote charge accumulation, but there is no universal 40% threshold and the mechanism and appearance vary by plate and reader.
Prevention: Keep temperature and humidity within the manufacturer-specified range, use approved handling and grounding practices, and use only manufacturer-approved cleaning products. Do not apply an antistatic spray to a PSP plate unless its instructions explicitly permit it.
A ghost image — a faint remnant of a prior exposure — appears when the PSP plate is not fully erased before reuse. CR plates store residual energy in the phosphor layer after readout; the erasure process (exposure to bright light) must completely release this remaining energy. If the erasure lamp is failing, the exposure time is insufficient, or the plate has been exposed to a very high dose, residual signal remains.
Residual signal can appear as faint superimposed anatomy or exposure-field edges from a prior image. Visibility depends on exposure, elapsed time, processing, and display polarity; a collimated, minimally exposed region is not universally a black background.
Prevention: Follow the manufacturer's erasure and plate-reuse instructions, including any special erasure after unusually high exposure or prolonged storage. Persistent carryover requires removal of the cassette/reader from clinical use as local policy directs and evaluation by service or medical physics; users should not alter service settings without authorization.
Use pattern recognition to form a hypothesis, not a diagnosis. Branching lines suggest static; regular parallel lines suggest a grid or reader/sampling problem; fixed irregular lines suggest plate or detector damage; and prior anatomy suggests CR incomplete erasure, double exposure, or DR lag. Dust on a plate may make a localized defect, while dust in a CR reader's light-collection path can make a line in the scan direction.
Digital radiography (DR) commonly uses flat-panel detectors. Indirect-conversion panels first convert x rays to light in a scintillator; direct-conversion panels convert x-ray energy to charge in a photoconductor. Fixed panels and wireless panels have different handling risks, and an artifact may arise from hardware, calibration, acquisition, or processing—not necessarily detector degradation.
An uncorrected defective detector element may appear at a fixed detector location as a bright or dark pixel, line, or cluster, depending on the defect and display polarity. Systems commonly use a manufacturer-specific defect map and interpolation from surrounding data, so many defective elements are not visible in a processed clinical image.
A newly visible fixed defect can reflect a changed element, calibration/map failure, electronics, or physical damage. Contiguous clusters and defective rows or columns are more difficult to conceal than isolated elements and may remove clinically relevant information.
Action: Identify and document the affected detector, preserve an example, and perform only the user QC specified by the manufacturer and facility. Defect-map updates and calibrations may be restricted to trained service or physics personnel. Acceptance limits are detector-, task-, and program-specific; there is no universal “10 pixels per 1 cm” rule.
DR detectors require a flat-field (gain) calibration to compensate for differences in pixel sensitivity across the detector surface. If the calibration is outdated or performed incorrectly, the resulting image shows subtle, large-area density variations — one region may appear slightly brighter or darker than another, even when imaging a uniform phantom. This is known as a "gain artifact" or "shading artifact."
Calibration can become invalid after hardware changes, contamination, exposure of the detector during calibration, or operation outside the manufacturer's thermal conditions. Depending on the fault, nonuniformity can appear as gradients, bands, seams, or retained outlines rather than only a gradual density change.
Action: Follow the vendor's warm-up and calibration instructions. A flat-field calibration must use the specified beam quality, geometry, field coverage, detector orientation, and exposure; an incorrectly made calibration can imprint objects or beam nonuniformity into later images. Persistent nonuniformity needs service or medical-physics evaluation.
Image lag is residual response from one acquisition that affects a later readout. It can arise from charge trapping, incomplete readout/reset, or scintillator/phosphor persistence, depending on detector design. A high detector exposure followed soon by a lower-exposure acquisition can make lag more conspicuous.
Lag is measured under defined timing and exposure conditions. There is no universal 5% clinical acceptance limit: specifications and action levels depend on detector technology, measurement method, frame timing, and imaging task.
Action: Follow specified acquisition timing and detector reset procedures. Recurrent visible carryover should be documented and evaluated with the manufacturer's or physicist's lag test; perform offset/dark or other calibration only when the approved procedure calls for it.
Anti-scatter grids improve contrast when scatter is substantial and the protocol calls for one, but they can introduce cutoff, visible-line, and alias artifacts. They are not required for every examination, and an unnecessary or misused grid can increase dose or reduce image quality.
Grid cutoff is loss of primary radiation caused by grid/tube/detector misalignment. It can be peripheral, one-sided, bilateral, or fairly uniform, depending on the error. Display processing can partly normalize global brightness but cannot restore the lost signal; affected regions can show increased quantum noise and reduced information.
Causes include lateral decentring of the central ray, tube or grid angulation across the lead strips, use outside the focal-distance range of a focused grid, a reversed focused grid, and damaged or incorrectly seated grids. A reversed focused grid classically causes severe peripheral cutoff while the central region is less affected.
A Moiré pattern is a broad banded or wavy alias pattern caused when a stationary grid's periodic structure is sampled inadequately. It is a digital sampling artifact and is especially well described with CR readers, although it can also occur with flat-panel systems.
Risk depends on grid-line frequency and orientation, detector or CR-reader sampling pitch in both directions, grid motion, and processing. Grid ratio does not determine the alias frequency. A periodic grid structure above the Nyquist limit can alias to a lower visible frequency, so saying that Moiré occurs only “close to Nyquist” is incomplete.
Solution: Use only a grid frequency and orientation approved for the detector or CR reader, verify that a moving grid moves correctly, and use validated grid-line-suppression processing where provided. Do not deliberately angle a focused grid as a generic fix because this can create cutoff. Merely choosing a grid above the detector's Nyquist frequency does not guarantee prevention.
Stationary grid strips may be resolved as fine parallel lines or may alias into broader bands. Some systems provide grid-suppression processing, but it is not universal and performance depends on compatible grids and correct configuration. A stopped moving grid, damaged grid, wrong grid selection, or processing failure can make lines conspicuous.
Solution: Inspect alignment and damage, verify moving-grid operation where applicable, and confirm the configured grid/protocol according to the manufacturer. Do not infer an undocumented “drifting algorithm”; recurring artifacts should be tested systematically.
Motion and positioning errors are not unique to digital imaging, and display processing does not restore spatial information lost during the exposure.
Motion during an exposure produces loss of edge sharpness and may create smeared or duplicated anatomy. The pattern depends on direction, amplitude, and exposure time. Whether a blur appears confined to anatomy is useful context but is not a definitive test of cause.
Respiration or gross movement can blur several structures, while cardiac, vascular, or bowel motion can be localized. Duplication may occur with abrupt movement, but no single “double-lung” appearance is specific for breathing.
Solution: Use an exposure time appropriate to the examination and equipment limits, clear breathing instructions, and safe positioning or immobilization consistent with policy. Raising mA to shorten time may be useful only within tube-loading, focal-spot, generator, AEC, and patient-dose constraints. Routine chest radiography is generally obtained at full inspiration unless the requested view or clinical condition requires another phase; end-expiration is not a general pediatric motion remedy.
Movement of a portable detector relative to the patient and beam can blur or displace projected anatomy. There is no universal sharp-border signature, and the collimated-border claim does not reliably distinguish detector from patient motion.
Solution: Stabilize the detector without placing unsafe pressure on the patient, avoid loading beyond the detector's specified limits, secure any holder, and ensure the patient and staff will not move it during exposure.
Insufficient detector exposure produces quantum mottle: random noise and loss of low-contrast information. Processing can make overall brightness look acceptable but cannot recover missing signal. At the other extreme, detector or analog-to-digital saturation can map a range of high exposures to the same maximum value, erasing information (“clipping”). Saturation behavior and warning indicators are system-specific; a bright or dark displayed region alone does not prove saturation.
Action: Review the unprocessed or “for processing” data only through authorized QC tools, the exposure indicator and deviation index, acquisition parameters, and any saturation warning. Do not repeat solely to place an indicator at its target if the image is diagnostically adequate. A repeat decision should follow departmental policy and clinical review; recurrent clipping or unexpected noise needs protocol, AEC, generator, or detector evaluation.
Scatter reaching the detector adds unwanted signal and reduces subject contrast; tight, anatomically appropriate collimation is the primary control, with an appropriate grid or air gap used when the protocol calls for it. A grid usually requires greater receptor exposure and, if technique is adjusted, can increase patient dose; it should not be added indiscriminately.
Radiation scattered from structures behind a CR cassette or portable DR panel can reveal cassette hinges, electronics, labels, bedding, or other support structures, particularly at high exposures or when shielding/backing is inadequate for the geometry. The pattern and polarity vary. Use the detector, cassette orientation, shielding, and positioning specified by the manufacturer—do not add improvised lead that could interfere with equipment or AEC.
Clothing, hair, skin folds, jewelry, monitoring leads, oxygen tubing, diapers, bed hardware, and objects under or behind the patient can obscure anatomy or mimic disease. Remove only items that can be removed safely, trace unavoidable lines/tubes when uncertainty exists, and never disconnect medical devices without authorization. Processing cannot reliably remove an external object from the diagnostic data.
One of the biggest challenges in digital radiography is that image processing algorithms can create, hide, or mimic artifacts. Understanding these processing artifacts is essential for the modern rad tech.
Many systems identify exposed fields and relevant values before applying anatomy-specific grayscale processing. Failure can follow poor or absent collimation, multiple exposures on one receptor, unusual positioning, a prosthesis or shielding, or selection of the wrong exam/protocol. Consequences include inappropriate brightness or contrast, segmentation lines, cropped anatomy, and an unreliable exposure indicator.
Solution: Collimate to the clinical area before exposure and select the correct exam and projection. Do not use electronic cropping as a substitute for beam collimation. Manual reprocessing or region-of-interest correction is vendor-specific and should preserve the original image and follow local policy; it must not conceal clipped anatomy or an acquisition error.
Digital display processing can conceal excessive detector exposure and allow exposure creep. Under the standardized IEC definition, DI = 10 log10(EI/EIT): DI 0 means the target detector exposure, a positive DI means EI exceeded target, and a negative DI means it was below target. EI estimates detector exposure, not patient dose; segmentation errors and beam quality can affect it, and some legacy/vendor indicators run in the opposite direction. Interpret the indicator only after confirming the system definition and target for that examination.
Digital radiography processing commonly includes frequency-dependent enhancement. Excessive enhancement can produce bright and dark halos (“overshoot,” “undershoot,” or ringing) near high-contrast boundaries; appearance depends on the algorithm and display polarity.
Excessive edge enhancement also amplifies image noise, giving the image a grainy or "digital" appearance. This is most noticeable in low-dose images where the noise floor is higher.
Solution: Use validated anatomy-specific processing. Reprocess from the preserved original data with an approved alternative when clinically appropriate; recurring halos or amplified noise require protocol review by the vendor/application specialist, radiologist, and medical-physics team rather than ad hoc user adjustment.
Aliasing occurs when sampled detail contains frequencies above the system's Nyquist limit; periodic structures such as stationary grid strips are a much more characteristic radiographic example than an ordinary diaphragm or heart border. Digital resampling, magnification, or display interpolation can also create jagged edges, but these should not automatically be blamed on detector sampling.
Solution: Use the detector, grid, acquisition mode, and image-processing protocol validated for the examination. Pixel binning and acquisition matrices are not user-selectable on every system, and a small focal spot changes geometric unsharpness rather than the detector's Nyquist frequency. Anti-alias filtering trades high-frequency response for reduced aliasing and should be configured and validated by the vendor or imaging-physics team.
Understanding the differences between CR and DR artifacts helps technologists identify the source of the problem and take appropriate corrective action.
| Artifact Type | CR Appearance | DR Appearance | Root Cause |
|---|---|---|---|
| Plate/surface damage | Fixed irregular line, spot, or patch; polarity varies | Panel impact or surface damage can also create fixed defects | CR: damaged/contaminated PSP; DR: panel or electronics damage |
| Static electricity | Possible branching “lightning” pattern | Not a typical sealed-panel artifact | Electrostatic discharge promoted by handling/dry conditions |
| Prior-image carryover | Prior anatomy or field edges | Prior-image residual (lag) | CR: incomplete erasure/double exposure; DR: detector persistence, charge trapping, or reset/readout behavior |
| Element/reader defect | Lines or bands can arise in the scanning/transport/optical path | Fixed pixel, cluster, row, column, or panel-seam defect | CR reader/plate fault; DR detector element, electronics, or correction-map fault |
| Grid lines or Moiré | Fine lines or broad alias bands; well recognized with stationary grids | Fine lines or alias bands are also possible | Incompatible grid frequency/orientation, inadequate sampling, stopped moving grid, or processing/configuration fault |
| Shading/nonuniformity | Plate, light-guide, laser, reader, or processing fault | Gradient, bands, seams, or retained calibration outline | Contamination, hardware, offset/gain calibration, beam, or thermal conditions |
| Dust/dirt | Localized plate defects or reader-direction lines | Surface contamination can obscure anatomy; internal fixed defects require service | Contaminated PSP/cassette, CR optical path, or detector cover |
| Backscatter | May reveal support structures, labels, hinges, electronics, or bedding; brightness/polarity varies | Radiation scattered from behind the receptor and inadequate system backing/shielding for the geometry | |
When you encounter an unexpected artifact, follow this step-by-step approach:
Do not improvise a “50% abdomen mAs” flash test. Uniformity and defective-element tests require specified beam quality, source-to-detector distance, field coverage, detector orientation, exposure level, image-for-processing access, analysis, and limits. Use the manufacturer's and facility's procedure, minimize staff exposure, compare with baseline, and record the result.
Many artifact risks can be reduced through appropriate acquisition, equipment care, and a documented QA programme.
A QA programme should define responsible staff, test methods, baselines, action levels, records, and escalation. Frequencies are set by regulation/accreditation, the manufacturer, a qualified medical physicist, workload, and performance history. AAPM TG-151 describes routine visual inspection and reject analysis plus periodic tests; it does not establish “daily flat field for every DR detector” or universal weekly plate cleaning/monthly erasure-lamp checks. Clean and calibrate only at the approved frequency and after relevant events or observed problems.
Prevent impact, fluid ingress, cable/connector damage, and loading beyond the rated distributed or point-load limit. Follow the detector's specified operating/storage temperature and humidity; there is no universal 15–30°C range. After transport or a major temperature change, allow the prescribed acclimation time to avoid condensation and invalid calibration. Use infection-control covers and transport protection as required without trapping contaminants or compromising heat dissipation.
Inspect grids and holders for damage, seating, contamination, and moving-grid function according to the QA schedule. Store and clean them as the manufacturer directs and protect them from impact or excessive loads. Replace a grid when damage, performance testing, or service criteria require it—not at an unsupported universal 5–7-year interval.
Use validated technique charts/AEC and anatomy-specific target EI ranges. Review exposure indicators as part of the whole image, not as pass/fail dose meters. Trends outside target require investigation of positioning, collimation/segmentation, body habitus, beam quality, AEC, protocol, and equipment before changing technique. Optimize with radiologist and medical-physics input; a normal-looking processed image does not by itself establish adequate image quality or appropriate patient dose.
Try these ARRT-style multiple choice questions based on this article. Click an option to check your answer — correct answers turn green, wrong ones turn red.