Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
IEC 62945 — Measuring Imaging Performance of X-Ray CT Security-Screening Systems
IEC 62945, published in September 2018, establishes standardized test methods and phantoms for evaluating the imaging performance of X-ray computed tomography (CT) security-screening systems used for baggage and parcel inspection at airports, government buildings, and critical infrastructure checkpoints. As security threats evolve and CT scanners become the gold standard for aviation security screening, this standard provides an essential framework for objectively measuring image quality metrics including spatial resolution, CT number accuracy, noise characteristics, and artifact presence.
CT security-screening systems must comply with both imaging performance standards (IEC 62945) and radiation safety standards (IEC 62423, IEC 60601 series) to ensure passenger and operator safety.
Key Imaging Performance Metrics and Test Methods
The standard defines a comprehensive suite of imaging performance evaluation procedures. The object length accuracy test verifies that the system correctly measures linear dimensions along the scan axis. The path-length CT value test assesses the accuracy of CT number (Hounsfield unit) measurements across different material thicknesses, which is critical for identifying threat materials by their attenuation characteristics. Noise Equivalent Quanta (NEQ) analysis provides a fundamental measure of image quality by quantifying the effective number of X-ray quanta contributing to each pixel — a metric that directly correlates with the scanner’s ability to discriminate between materials with similar densities.
| Test Procedure | Metric | Test Article | Primary Purpose |
|---|---|---|---|
| Object length accuracy | Length measurement error (mm) | Precision rod assembly | Verify dimensional accuracy along scan axis |
| Path-length CT value | CT number vs. material thickness | Step wedge phantom | Assess CT number linearity with thickness |
| Noise Equivalent Quanta (NEQ) | Effective photon count per pixel | Uniform water-equivalent phantom | Fundamental image quality metric |
| CT value consistency | CT number stability across scans | Multi-material phantom | Verify calibration drift over time |
| CT value uniformity | Spatial variation of CT numbers | Uniform phantom | Detect beam hardening and scatter artifacts |
| Streak artifacts | Artifact severity score | Pin array phantom | Evaluate metal-induced artifact mitigation |
| Slice sensitivity profile (SSP) | Z-axis resolution (mm FWHM) | Slanted edge phantom | Characterize slice thickness and aliasing |
| Image registration | Spatial alignment accuracy | Registration test object | Verify geometric fidelity of reconstructed images |
Engineering Insight: NEQ as a Universal Image Quality Benchmark
The NEQ metric defined in IEC 62945 is particularly valuable because it encapsulates multiple imaging chain factors — detector quantum efficiency, electronic noise, scatter rejection, and reconstruction algorithm performance — into a single physically meaningful number. For system designers, tracking NEQ under varying exposure parameters provides a direct path to optimizing the radiation dose versus image quality trade-off. A practical target for baggage CT systems is an NEQ above 2,000 photons per pixel at the center of a 30 cm water-equivalent phantom, which typically correlates with reliable threat detection performance.
Test Article Specifications and Phantom Design
Annex A of IEC 62945 provides detailed engineering drawings and material specifications for two primary test articles (Test Article A and Test Article B) and several specialized phantoms. These test articles are designed to be manufacturable from commercially available materials with tight tolerances. Test Article A serves as a multi-purpose phantom incorporating features for assessing CT number accuracy, uniformity, and streak artifacts. Test Article B focuses on spatial resolution and registration accuracy. The standard specifies that all test articles must be made from materials that are stable over time, non-hygroscopic, and have well-characterized X-ray attenuation properties — typically polymethyl methacrylate (PMMA), water-equivalent plastics, and aluminum alloys for reference markers.
Using standardized phantoms from IEC 62945 ensures that imaging performance measurements are comparable across different CT security-screening systems from various manufacturers, enabling objective procurement evaluations.
Design Recommendation: Automated Phantom Positioning
For high-throughput security environments where daily performance verification is required, consider integrating an automated phantom positioning system that can place the test article at the isocenter with sub-millimeter repeatability. Manual positioning introduces operator-dependent variability that can mask subtle performance degradation. A motorized phantom stage controlled by the system’s quality assurance software reduces measurement uncertainty and frees screening personnel for security-critical tasks.
Statistical Analysis and Compliance Evaluation
Annex C of the standard provides statistical guidance for comparing CT system performance between baseline and candidate configurations. Two scenarios are addressed: comparing a single CT system after modification against its own baseline, and comparing a candidate system against an existing historical population. The standard recommends using Student’s t-test for paired comparisons and provides confidence interval calculations for determining whether observed performance differences are statistically significant. This statistical framework is essential for maintenance decisions — for instance, determining whether a detected change in NEQ warrants tube replacement or detector recalibration versus being attributable to normal measurement variation.
| Scenario | Statistical Method | Sample Size Recommendation | Decision Threshold |
|---|---|---|---|
| Single system post-modification vs. baseline | Paired t-test | 10 scans minimum | 95 % confidence interval |
| Candidate vs. historical population | Two-sample t-test | 10 candidate + 20 historical | p < 0.05 |
| Daily quality assurance trend | Shewhart control chart | 20 baseline measurements | 2σ warning / 3σ action |
Statistical significance does not always imply operational significance. A CT value shift of 5 HU may be statistically significant with a large sample size but have negligible impact on threat detection performance. Always pair statistical analysis with practical image quality assessment.
Frequently Asked Questions
How often should the full IEC 62945 test suite be performed on an operational CT security scanner?
The standard recommends a tiered approach: daily quick checks using a simplified phantom to verify CT number stability and object length accuracy, monthly comprehensive tests covering NEQ and streak artifacts, and annual full characterization including all test procedures specified in Clause 4. More frequent testing may be warranted after tube replacement, detector maintenance, or major software upgrades.
Can IEC 62945 test methods be applied to dual-energy or multi-energy CT scanners?
Yes, with appropriate adaptation. The standard’s test methods for CT number consistency and uniformity can be applied to each energy bin independently. The material discrimination performance inherent to dual-energy systems is not directly addressed by IEC 62945 and should be evaluated using supplementary test protocols specified by the manufacturer or regulatory authority.
What is the relationship between IEC 62945 and ASTM F3094?
ASTM F3094 also addresses CT security-screening system performance testing. IEC 62945 is more comprehensive in its coverage of image quality metrics (including NEQ, which ASTM F3094 does not specify) and provides more detailed phantom drawings and statistical guidance. Users in jurisdictions adopting IEC standards typically reference IEC 62945, while those following ASTM guidelines reference F3094. The two standards share some common test philosophies but differ in their specific phantom designs and pass-fail criteria.
How should environmental conditions be controlled during performance testing?
Clause 5 specifies environmental requirements: temperature within 15 °C to 35 °C, relative humidity below 80 %, and no direct sunlight on the detector array. The system must be warmed up according to the manufacturer’s specifications before testing. These conditions ensure that measured performance variations reflect true system capability rather than environmental artifacts.
