The Transportation Security Administration (TSA), a component of the Department of Homeland Security (DHS) responsible for the security of the transportation systems in the United States, has deployed security systems of advanced imaging technology (AIT) to screen passengers at airports. According to TSA, AIT systems provide enhanced security benefits by detecting both metallic and nonmetallic threat items, including weapons, explosives, and other concealed objects on passengers that would not be detected by walk-through metal detectors. To date (December 2014) TSA has deployed AITs in U.S. airports of two different technologies: millimeter wave and X-ray backscatter AIT systems. These technologies use different types of radiation to penetrate through clothing and create a reflected image of the body. Millimeter wave AIT systems use radio waves, a type of nonionizing radiation, while X-ray backscatter AIT systems use X-rays, a type of ionizing radiation. This report focuses on the X-ray backscatter AIT used by TSA for screening passengers.
DHS requested that the National Research Council (NRC)1 provide an examination of the radiation exposures resulting from X-ray backscatter AIT used in screening travelers. This examination would include an analysis of whether
1 Effective July 1, 2015, the institution is called the National Academies of Sciences, Engineering, and Medicine. References in this report to the National Research Council are used in a historical context identifying programs prior to July 1.
- The X-ray backscatter AIT system that uses ionizing radiation complies with applicable health and safety standards for public and occupational exposures to ionizing radiation, and
- AIT system design, operating procedures, and maintenance procedures are appropriate to prevent over exposures of travelers and operators to radiation.
The complete statement of task for the NRC Committee on Airport Passenger Screening: Backscatter X-Ray Machines is shown in Appendix A.
In 2008, TSA deployed the X-ray backscatter AIT systems in airports, specifically Rapiscan Secure 1000 units manufactured by Rapiscan Systems. TSA subsequently removed the Rapiscan Secure 1000 units from all airports by June 2013 because of the manufacturer’s inability to develop privacy software (known as automatic target recognition, or ATR), which enabled the display of anomalies on a generic figure instead of an image of the individual’s body, by the deadline mandated by Congress. At the time the committee started its examination, there were no X-ray backscatter AIT systems deployed in airports in the United States. The only AIT systems deployed at airports for passenger screening at the time of the committee examination were the millimeter wave units. TSA is currently evaluating second-generation X-ray backscatter AITs manufactured by American Science and Engineering, Inc. (AS&E).
Despite removing the Rapiscan Secure 1000 units from all airports, DHS expressed its interest in the NRC performing an examination of the Rapiscan Secure 1000 that was previously deployed in U.S. airports. Because there is a possibility that TSA could deploy second-generation X-ray backscatter AITs manufactured by AS&E, DHS asked NRC to also examine an AS&E AIT system. The AIT systems available to the NRC committee for this examination were a Rapiscan Secure 1000 unit located at the National Institute of Standards and Technology and an AS&E SmartCheck prototype located at TSA’s Transportation Systems Integration Facility. The two AIT systems available to the committee represent past (Rapiscan Secure 1000) and possible future (AS&E SmartCheck) X-ray backscatter AIT systems for airport passenger screening. The committee did not examine other X-ray backscatter AIT systems available today, such as Tek84’s Ait84. The Ait84 system is not equipped with ATR and, therefore, is not a candidate for deployment at U.S. airports.
In its examination of the Rapiscan Secure 1000 and AS&E SmartCheck systems, the committee was specifically asked not to evaluate the justification of the use of X-ray backscatter AIT systems, although an alternative technology that does not use ionizing radiation exists. The committee was also asked not to evaluate whether the standard with which this technology needs to comply is appropriate and/or
adequate to protect human health. The committee was asked if the AITs conform to applicable standards.
X-ray backscatter AITs must conform to the American National Standards Institute/Health Physics Society (ANSI/HPS) Accredited Standards Committee N43 (Equipment for non-medical radiation applications) N.43.17-2009 standard.2 The ANSI/HPS N43.17-2009 standard provides guidelines for both manufacturers and users of AIT systems and covers dose to subject, interlocks, operational procedures, and information to be provided to the travelers by the operators. The standard limits the dose per screening to 250 nanosievert (nSv) (0.25 µSv), called the reference effective dose. Similarly, the annual dose received by an individual from one facility shall not exceed 250,000 nSv (250 µSv). The reference effective dose, a simplified version of the effective dose, is easier to calculate but considered sufficiently accurate to ensure radiation safety. The standard does not issue different limits for subgroups of the general population, such as pregnant women and children, implying that the dose limit offers sufficient protection to them also.
To address its statement of task, the committee performed a comprehensive analysis of previous investigations of the AITs (see Chapter 6). In addition, the committee directed the work performed by an NRC-contracted team (see Chapter 7) tasked with making dosimetry and beam characteristics measurements for both the Rapiscan Secure 1000 and AS&E SmartCheck AITs. A second NRC-contracted team performed detailed computations to screened passengers using Monte Carlo methods with a suite of digitized human phantoms (see Chapter 7).
Results of the subcontractors’ work were used by the committee to draw its findings and recommendations for the first part of the statement of task related to whether exposures from the AIT systems comply with the applicable health and safety standard. The committee was unable to carry out AIT system design reliability analyses because this material is proprietary and is protected from public release. Therefore, to respond to the second part of the statement of task, the committee inspected and tested some AIT safety systems and, through the NRC subcontractors, performed a detailed failure mode analysis of the dose received by the person being screened in situations where safety interlocks fail and overexposure may be possible.
Overall, the committee found that both the Rapiscan Secure 1000 and AS&E SmartCheck X-ray backscatter AIT systems complied with the applicable standard. This finding was in agreement with previous investigations of the doses received under routine operation of the X-ray backscatter AITs. The committee also found that the dose received under worst-case scenarios of AIT system failure, those
2 The ANSI/HPS N43.17-2009 standard, “Radiation Safety for Personnel Security Screening Systems Using X-Ray or Gamma Radiation,” is available at the Health Physics Society website at http://hps.org/hpssc/index.html.
failures that result in the beam becoming stuck while still producing X rays, does not exceed the applicable standard and therefore cannot result in overexposure.
The key findings and recommendations that emerged from the committee’s examination are presented and discussed below. Additional technical details to support these key findings and recommendations are presented in Chapters 6 and 7 of the report.
KEY FINDINGS AND RECOMMENDATIONS
Dose Reported in Previous Investigations
Individual Being Screened
Key Finding: Previous radiation dose studies employed different methodologies and instrumentation to estimate the dose delivered by Rapiscan Secure 1000 X-ray backscatter AITs. These studies generally agreed that the radiation exposure dose per screening to an average passenger is about a factor of 10 below the limit of 250 nSv per screening, the dose set by the ANSI/HPS N43.17 2009 standard.
Individual Outside the Screening Area
Key Finding: Measurements of radiation outside the inspection area have generally been made with detectors that are calibrated for X-ray energies higher than those scattered from a passenger being scanned by the Rapiscan Secure 1000 system. As a result, the detectors may indicate a lower dose than is actually present. The detectors have sometimes failed to distinguish a signal from the background radiation.
Key Recommendation: To estimate X-ray radiation exposure outside the inspection area, measurements should be made with detectors calibrated for X-ray energies below the maximum for the AIT system’s X-ray tube and for the radiation levels expected. Use of detectors that are appropriate for other applications but not ideal for measuring dose in a X-ray backscatter AIT system may result in inaccurate measurements.
Dose Reported in the NRC-Contracted Study
Individual Being Screened
The committee’s approach in examining dose to the individual being screened differs from that of previous investigations in two ways:
- It made use of sensitive detectors with tissue-equivalent phantoms to verify beam intensity, X-ray quality, and penetration; and
- It performed computations using estimates of beam intensity, scanning geometry, and digitized (high-fidelity) human phantoms that have realistic dimensions and morphology.
Based on these improvements in approach, the committee has the following key finding:
Key Finding: For either the Rapiscan Secure 1000 or the AS&E SmartCheck systems, as determined by the committee for adults and children,
- The effective doses are about the same as those calculated following the simplified formula for the reference effective dose identified by the ANSI/HPS N43.17-2009 standard;
- The effective doses are lower than those in previous reports using plane-parallel X-ray beams with stylized geometrical (low-fidelity) human phantoms; and
- Sensitivity analysis showed that under a range of different conditions, including passenger position in the AIT system and increases in the energy (i.e., by increasing the tube high voltage) of the X-ray beam, the computed effective dose would not increase by more than a factor of 3 and, even so, would remain below the limit specified in the ANSI/HPS N43.17-2009 standard.
The committee can also make the following more specific statements:
- No person, regardless of age and weight modeled, would exceed the effective dose limit per screen (i.e., 250 nSv/screen), as defined by the ANSI/HPS N43.17-2009 standard.
- The absorbed dose per screen to the developing fetus at any of the three stages post-conception is less than 0.0002 percent of the recommended limit for radiation protection of the fetus during the entire gestation period.
- The absorbed dose to the epithelial layer of radiosensitive cells in the skin is not significantly elevated (~1.6 percent) compared to the average absorbed dose to the skin.
- The dose received by the lens of the eye, skin, or female breast during a stationary beam of X rays for the duration of the scan were at least 2 orders of magnitude below thresholds where tissue injury might occur.
In general, the committee found the following:
Key Finding: Under routine operations, the computed effective doses using computational X-ray sources and scanning geometries, coupled with the digitized hybrid phantoms, are similar to the ANSI reference effective dose and an order of magnitude below the limit of 250 nSv/screen, as set forth in the applicable ANSI/HPS N43.17-2009 standard.
Key Finding: The agreement between the estimated dose results from the NRC subcontractor and the results from earlier studies confirms that the reference effective dose calculations performed in the previous studies were adequate to establish compliance with effective dose limits recommended in the ANSI/HPS N43.17-2009 standard.
Individual Outside the Screening Area
Key Finding: Using appropriate detectors, the estimated values of the radiation outside the inspection area that might affect a bystander are so low as to be statistically indistinguishable from background radiation.
AIT System Design Evaluated in the NRC-Contracted Study
Although the NRC subcontractors performed detailed radiation measurements and dose computations, the committee was unable to unequivocally determine whether the X-ray backscatter AIT systems studied have adequate operating safety interlocks that would prevent the AIT system from exceeding the ANSI/HPS N43.17-2009 standard under every imaginable situation. This is because
- The committee was not given an opportunity to independently verify how all of the interlocks perform in different situations, with the exception of
- simple functions such as termination of operation if a door was opened. Such testing would need engineering support from the manufacturer and require unique testing tools as well as dismantling portions of the AIT systems, which would potentially cause damage.
- The committee was not given a demonstration of how interlocks are checked at the manufacturer level from either Rapiscan or AS&E.
- Detailed electrical and mechanical drawings and computer code descriptions and documents describing internal functions at the most fundamental level of the AIT systems are either restricted from public access or were not made available by manufacturers to either the committee or the sponsor.
However, the committee was able to inspect the interior of both the Rapiscan Secure 1000 and AS&E SmartCheck AIT systems. AS&E representatives also described to the committee how many of the interlocks are intended to perform on their second-generation prototype.
With the above limitations noted, having evaluated as many aspects of the AIT systems’ safety interlocks as possible, both mechanical and electrical, and combining that knowledge with the measurements and computations performed, the committee can make the following statements:
Key Finding: It appears that the X-ray backscatter AIT systems adhere to the recommended safety mechanisms described in the ANSI/HPS N43.17-2009 standard.
Key Finding: Acceptance tests and periodic inspection tests guided by the safety inspection forms previously used during deployment are sufficient to meet the indicators, controls, and safety interlocks requirements of the ANSI/HPS N43.17-2009 standard.
Key Finding: Equipment manufacturers recommend that a test piece be scanned daily to evaluate proper operation of the AIT system, because this ensures that many of the needed safety system requirements in ANSI/HPS N43.17-2009 standard work properly. The committee agrees with this recommendation but was unable to determine if this was done, because the X-ray backscatter AIT systems are not currently deployed in the field at commercial airports.
Key Recommendation: Any future testing procedures should at a minimum continue to follow the indicators, controls, and safety interlocks requirements of the ANSI/HPS N43.17-2009 standard, or similar testing procedures, and include daily verification of safety parameters by a test piece.
Key Finding: Based on the committee’s review and test of the Rapiscan Secure 1000 system’s interlocks, the committee was unable to identify any circumstances where an accidental failure or deliberate reconfiguration of the AIT system could result in either a person being screened or the operator receiving an effective dose larger than that from a normal screening.
Key Finding: Based on the committee’s inspection of the AS&E SmartCheck AIT system, with the AS&E representative present, the committee was unable to identify any circumstances where an accidental failure or deliberate reconfiguration of the AIT system could result in either a person being screened or the operator receiving a larger X-ray dose than the intended screening dose.
Key Finding: Given the results obtained by the committee on radiation measurements and computations for the X-ray AIT systems investigated, the screening of an individual would need to extend for more than 60 seconds for an individual to be exposed to radiation that exceeds the ANSI/HPS N43.172009 standard limit. In comparison, a typical screen takes about 6 seconds.
Key Recommendation: Future X-ray AITs should have some independent mechanism to ensure that the AIT does not screen any person for longer than the time needed to acquire the appropriate image while keeping radiation exposure compliant with the safety principle of as low as (is) reasonably achievable.