Does a CT room need a primary barrier?

No. A CT scanner has no fixed useful beam that escapes the gantry to strike a wall, floor, or ceiling. The fan beam rotates continuously around the patient and is absorbed by the patient and the gantry assembly, so there is no primary beam direction to shield against. Every CT barrier is therefore a secondary barrier designed against scattered radiation and a small amount of tube-housing leakage. This is the key difference from a radiographic or fluoroscopic room, which does require primary barriers where the collimated beam can strike a wall.

What is the dose-length-product (DLP) method for CT shielding?

The DLP method in NCRP Report No. 147 estimates the scattered air kerma around a CT scanner from the facility's weekly workload expressed as total dose-length product. The scattered air kerma at one meter from the isocenter is taken as a scatter factor, often written as the Greek letter kappa, multiplied by the weekly DLP. NCRP 147 gives representative scatter factors of roughly 9 by 10 to the minus five per centimeter for head technique and 3 by 10 to the minus four per centimeter for body technique at one meter. Multiplying by the weekly DLP and correcting for distance gives the unshielded weekly air kerma at each occupied location.

What design goals apply to a CT installation?

NCRP Report No. 147 sets shielding design goals, P, as a weekly air kerma on the occupied side of a barrier: about 0.1 mGy per week for controlled areas (5 mGy per year) and about 0.02 mGy per week for uncontrolled areas (1 mGy per year). These design goals are deliberately more conservative than the regulatory dose limits in 10 CFR Part 20, which cap public exposure at 1 mSv per year and occupational exposure at 50 mSv per year. Most spaces around a CT suite are uncontrolled areas, so the 0.02 mGy per week goal usually governs.

Why are manufacturer isodose scatter maps used in CT shielding?

Each CT vendor publishes isodose scatter maps that plot the scattered air kerma per scan, or per unit of CTDI or DLP, at measured points in the horizontal and vertical planes around the gantry. These maps capture the real spatial distribution of scatter, including the shadowing effect of the gantry and couch, which the simpler DLP point-estimate cannot. NCRP 147 accepts the isodose-map method as an alternative, and comparison studies have found it the most spatially reliable of the available methods, particularly at points beyond and behind the gantry.

How much lead does a typical CT room need?

It depends entirely on workload, distance, and occupancy, but a representative busy adult body-CT room often requires on the order of 1 to 2 mm of lead (roughly 1/16 inch) on walls adjacent to fully occupied uncontrolled areas, with less where distance or low occupancy helps. Concrete or gypsum alternatives are converted from the required transmission using the broad-beam attenuation data in NCRP 147. The only way to know the true requirement is a documented calculation for the specific room and adjacencies, not a rule of thumb.

Is a survey required after a CT room is built?

Yes. A post-construction radiation protection survey, performed by a qualified medical physicist, verifies that the installed barriers actually achieve the design goals before clinical use. The physicist operates the scanner under representative technique and measures scattered air kerma at occupied locations on the far side of each barrier, checking for gaps at joints, doors, conduit penetrations, and viewing windows. Many state radiation-control programs require this survey and a shielding plan review as conditions of registration.

CT Room Shielding Design: Scatter & Leakage

A CT scanner room is a secondary-barrier problem: there is no primary barrier, because no useful beam ever escapes the gantry to strike a wall. The rotating fan beam is absorbed by the patient and the gantry assembly, so the only radiation reaching the barriers is scattered radiation from the patient plus a small amount of tube-housing leakage. That single fact separates CT shielding from radiographic and fluoroscopic shielding and reshapes the entire calculation—the workload is expressed in CT output terms, the dominant source is patient scatter, and every barrier is sized from the scatter and leakage transmission alone.

This guide explains the CT-specific shielding methodology of NCRP Report No. 147, works a representative barrier calculation using the dose-length-product scatter method, and describes the isodose-map alternative and the post-construction survey that confirms the design. DRPS provides this analysis as part of its radiation shielding design service.

Introduction

The central question in CT shielding is not where the beam points, but how much the patient scatters. In a radiographic room, a collimated useful beam strikes a defined wall or floor, so a thick primary barrier must sit where the beam lands, and thinner secondary barriers handle scatter and leakage elsewhere. A CT scanner works differently. The x-ray tube rotates continuously around the patient while a narrow fan or cone beam is collimated to the detector arc on the opposite side of the gantry. The patient absorbs most of the beam, the gantry hardware intercepts the rest, and no collimated useful beam leaves the gantry aperture toward a room surface.¹⁶

The result is that CT rooms are designed entirely with secondary barriers. The two sources that reach the walls are, first, radiation scattered from the patient in every direction, and second, leakage radiation that penetrates the tube housing. Of these, patient scatter dominates by a wide margin; leakage is held to a low limit by the tube-housing performance standard and is usually small compared with scatter at diagnostic CT techniques.¹²

Because scatter is the design driver, the workload cannot be expressed the way a radiographic room's workload is. NCRP 147 ties CT scatter to the scanner's output, quantified by the computed tomography dose index and the dose-length product, rather than to tube-current-exposure-time products delivered toward a primary wall.¹ The modern CT installation is high-output and high-throughput, so a busy scanner can deliver substantial weekly scatter, and an adjacent full-time office or control area can approach the uncontrolled-area design goal if the barrier is not sized correctly.⁷⁸

Topic Explanation

Primary versus secondary barriers—and why CT has only one kind

A primary barrier is designed to attenuate the useful (primary) beam. A secondary barrier is designed to attenuate scattered radiation and leakage radiation. In a general radiographic or radiographic-fluoroscopic room, both exist: the chest-board wall and the floor under the table are primary barriers, while the control booth and side walls are secondary barriers.³

A CT scanner produces no barrier-bound primary beam. The fan beam is confined within the gantry, aimed at the detector ring, and attenuated by the patient. Consequently, NCRP 147 treats the CT room as an entirely secondary-barrier installation.¹ This is the single most important conceptual point for anyone reviewing or sizing a CT barrier, and it is the clean distinction from the general diagnostic X-ray room shielding under NCRP 147 approach, which must place primary barriers where the collimated beam can land.

Radiation component	Radiographic / fluoroscopic room	CT room
Useful (primary) beam	Primary barrier required on beam-on surfaces	Not applicable—beam contained in gantry; no useful beam reaches a barrier
Patient scatter	Secondary barrier	Primary design driver (secondary barrier)
Tube-housing leakage	Secondary barrier	Secondary barrier (small relative to scatter)

Workload expressed as CT output

For a radiographic room, workload is built from the weekly number of exposures and the tube loading (mAs) directed at each barrier. For CT, there is no barrier-directed loading, so NCRP 147 expresses workload through CT output descriptors: the weighted or volume computed tomography dose index (CTDI) and, most usefully for shielding, the dose-length product (DLP), which is the volume CTDI multiplied by the scan length.¹² The facility's weekly workload becomes the sum of the DLP over all scans performed in a week, separated where appropriate into head and body technique because the two produce different scatter per unit output.⁶

The DLP captures both how much radiation the scanner delivers per rotation and how long an anatomy is scanned, which is exactly what governs the total scatter emitted into the room. This is why a DLP-based or mAs-based weekly workload—not a primary-beam workload—is the correct starting point for a CT barrier.¹

Controlled versus uncontrolled areas

A CT suite is surrounded mostly by uncontrolled areas: reception, waiting rooms, corridors, adjacent offices, and any occupied space above or below. NCRP 147 assigns these a lower weekly design goal than controlled areas such as the CT control room, where occupationally monitored staff work. Because uncontrolled spaces usually surround the scanner, their lower design goal normally governs the thickest barriers.¹³

Key Technical Principles

The shielding design goals

NCRP 147 expresses the shielding design goal, P, as a weekly air kerma that must not be exceeded on the occupied side of a barrier. These goals are design targets that already embed conservatism; they sit below the regulatory dose limits of 10 CFR Part 20, which limit public dose to 1 mSv per year and occupational dose to 50 mSv per year.¹⁴

Area type	Weekly design goal P	Annual basis
Controlled area (monitored workers)	0.1 mGy/week	5 mGy/year
Uncontrolled area (public / unmonitored)	0.02 mGy/week	1 mGy/year

Occupancy and distance

Two modifiers scale the raw scatter before it is compared with the design goal. Distance reduces air kerma by the inverse-square law, so a barrier 4 m from the isocenter sees one-quarter of the scatter that a barrier 2 m away sees. Occupancy, the factor T, is the fraction of the week that the most-exposed person actually spends in the adjacent area; a barrier is only required to protect a person who is present.¹³

Area (representative, NCRP 147)	Occupancy factor T
Offices, labs, reception, nurse stations, occupied adjacent rooms	1
Adjacent CT control room, patient exam rooms	1/2
Corridors, employee lounges, staff rest rooms	1/5
Corridor doors	1/8
Public toilets, vending, storage, unattended waiting, outdoor seating	1/20
Unattended parking, stairways, attics, transit-only outdoor areas	1/40

The scatter / dose-length-product method

NCRP 147's DLP method estimates the unshielded scattered air kerma at an occupied location directly from the weekly workload. The scattered air kerma at a reference distance of 1 m from the isocenter is the product of a scatter factor κ (kappa) and the weekly DLP:

\dot{K}_{1} = κ N \overline{DLP}

where $\dot{K}_{1}$ is the weekly scattered air kerma at 1 m (mGy·week⁻¹), $κ$ is the scatter air kerma per unit DLP at 1 m (mGy per mGy·cm), $N$ is the number of scans per week, and $\overline{DLP}$ is the mean dose-length product per scan (mGy·cm). NCRP 147 gives representative body-technique scatter factors near $κ \approx 3 \times 1 0^{- 4}$ mGy·(mGy·cm)⁻¹ at 1 m and head-technique factors near $9 \times 1 0^{- 5}$ ; independent measurements by Wallace and colleagues reported comparable values of about 0.36 µGy per mGy·cm for body and 0.14 µGy per mGy·cm for head scans at 1 m.¹⁶

Correcting to the actual distance d (meters) from the isocenter to the occupied point gives the unshielded weekly air kerma there:

K_{unshielded} (d) = \frac{κ N DLP}{d ^{2}}

The required barrier transmission B is the ratio of what is allowed to what arrives, scaled by occupancy:

B = \frac{P}{T K _{unshielded} ( d )}

Finally, B is converted to a material thickness. The rigorous route is the Archer three-parameter transmission model, the empirical broad-beam fit adopted by NCRP 147:

x = \frac{1}{α γ} ln \frac{B ^{- γ} + \frac{β}{α}}{1 + \frac{β}{α}}

where x is the barrier thickness and α, β, and γ are tabulated fitting parameters for a given material and scattered-beam spectrum (α is the slope at large thickness, β the slope at small thickness, and γ the curvature).⁵ A convenient approximation, adequate for a first estimate, converts B to a number of half-value layers:

n = \frac{lo g _{10} ( 1/ B )}{lo g _{10} 2}, x_{Pb} = n \cdot HVL_{Pb}

Worked example

Consider a busy adult body-CT scanner performing N = 120 scans per week at a mean $\overline{DLP}$ = 500 mGy·cm, with a fully occupied uncontrolled office sharing a wall 2.5 m from the isocenter (T = 1).

Weekly scatter at 1 m, using κ = 3 × 10⁻⁴ mGy·(mGy·cm)⁻¹:

\dot{K}_{1} = (3 \times 1 0^{- 4}) (120) (500) = 18 mGy/week

Unshielded air kerma at the office wall, 2.5 m away:

K_{unshielded} = \frac{18}{( 2.5 ) ^{2}} = 2.88 mGy/week

Required transmission against the uncontrolled goal P = 0.02 mGy/week:

B = \frac{0.02}{( 1 ) ( 2.88 )} = 6.9 \times 1 0^{- 3}

Converting to lead by the HVL approximation, with a representative CT-scatter lead HVL of about 0.2 mm at 120 kVp:

n = \frac{lo g _{10} ( 1/6.9 \times 1 0 ^{- 3} )}{lo g _{10} 2} \approx 7.2 HVL, x_{Pb} \approx 7.2 \times 0.2 \approx 1.4 mm

This rounds up to the next standard lead sheet of 1/16 inch (1.6 mm), a typical result for a busy CT wall against a full-occupancy neighbor.¹⁵ The Archer model applied to NCRP 147's broad-beam curves refines this value and is what a physicist uses for the final specification; the HVL estimate is only a sanity check. Note that the DLP method is deliberately conservative—it credits no attenuation from the gantry or couch, which can intercept a large fraction of scatter in some directions—and it has been shown in some geometries to underestimate measured scatter, so physicists add margin and cross-check against isodose maps.⁸

The isodose-map method

Every CT vendor supplies isodose scatter maps: measured contours of scattered air kerma, normalized per scan or per unit CTDI/DLP, plotted at points in the horizontal and vertical planes around the gantry. Because they are measured on the actual machine, they capture the true spatial distribution of scatter—including the shielding effect of the gantry and couch and the reduced scatter directly fore and aft of the bore—that the single-point DLP estimate cannot. NCRP 147 accepts the isodose-map method as an alternative, and a comparison of four CT shielding methods using the RadShield tool found the isodose-map approach the most spatially reliable, especially at points beyond and behind the gantry, while overall barrier recommendations across methods were broadly similar.⁵⁷ A practical caution: manufacturer scatter data have on occasion been found to overstate or understate measured scatter, so the physicist verifies the design against measurement rather than trusting a single data source.⁹

Clinical Impact

Correct CT shielding protects three groups. Members of the public in adjacent uncontrolled spaces—patients in waiting rooms, staff in neighboring offices, people on floors above and below—are kept below the 1 mGy per year design basis. Occupationally exposed staff, chiefly the technologists in the control room, are kept well within occupational limits and as low as reasonably achievable. And the facility itself is protected from the cost and disruption of discovering, after construction, that a barrier is inadequate and must be retrofitted behind finished walls.¹⁴

The stakes have risen with modern scanner performance. Multi-detector CT delivers high output and sustains high patient throughput, and intra-operative and interventional CT can place staff near the gantry during acquisition. Scatter from a wide-beam multi-detector scanner is substantially higher per scan than from the single-slice machines of a generation ago, scaling roughly with beam width, so shielding adequate for an older scanner is not automatically adequate for its replacement.⁹ When a facility swaps a scanner, the shielding assumptions must be re-examined against the new unit's output and isodose data.

Practical Optimization Tips

Start from a realistic weekly workload. Use the facility's actual or projected scan mix and DLP, separated into head and body technique, rather than a generic number. Over-estimating workload wastes lead; under-estimating it fails the survey.¹
Exploit distance before buying lead. Inverse-square is free attenuation. Where the floor plan allows, position the gantry isocenter away from the most sensitive adjacency, or place low-occupancy spaces (storage, mechanical rooms) against the busiest barriers.³
Apply honest occupancy factors. A corridor or restroom genuinely warrants a reduced T; a full-time workstation does not. Document the basis for each factor so a reviewer can follow it.¹
Design for the real adjacencies in three dimensions. Floors and ceilings are barriers too. Scatter travels up and down, so an occupied space directly above or below the scanner needs the same analysis as a side wall.¹
Use the manufacturer's isodose maps, but verify them. Cross-check the DLP estimate against the vendor's isodose data and reconcile differences; treat a large discrepancy as a reason to measure, not to pick the smaller number.⁷⁹
Shield the weak points, not just the walls. Doors, control-room viewing windows, cable and conduit penetrations, and construction joints are where barriers fail. Lead-lined doors, leaded glass of matching lead equivalence, and overlapping lead at seams preserve the designed transmission.³
Build in margin for the next scanner. Specifying the wall for a plausible future higher-output unit is far cheaper than opening finished walls later.⁹

Regulatory Considerations

CT scanners are x-ray-producing devices, not radioactive material, so they fall under a different regulatory structure than nuclear-medicine sources. At the federal level the U.S. Food and Drug Administration sets the electronic-product performance standard for CT equipment at 21 CFR 1020.33, which governs the scanner's output characteristics and defines CTDI, and the tube-housing leakage limit that bounds the leakage component of a secondary barrier.¹ The NRC's 10 CFR Part 20 does not apply to the x-ray machine itself; it establishes the dose limits for radioactive material that underlie the design goals, but a CT unit is regulated by the FDA and by the state, not by the NRC.⁴

Day-to-day oversight of CT installations—registration of the machine, review of the shielding plan, and inspection—rests with state radiation-control programs, whose requirements derive from NCRP 147 and the Suggested State Regulations. Many states require that a shielding plan be prepared or reviewed by a qualified medical physicist before construction, and that a post-construction survey confirm the installed barriers before clinical use. In Florida, for example, x-ray machines including CT are registered and inspected by the Florida Department of Health, Bureau of Radiation Control, under Chapter 64E-5, F.A.C., Part V; because Florida is an NRC Agreement State, the state—not the NRC—carries out this review and inspection.¹ A facility must confirm its own state's plan-review and survey rules, because they vary.

NCRP Report No. 147 (2004) is the governing structural-shielding report for medical imaging x-ray facilities and remains the current edition; it superseded the older NCRP Report No. 49 and provides the design goals, workload concepts, occupancy factors, and transmission data used throughout this analysis.¹¹¹

Frequently Asked Questions (FAQs)

Why does a CT room need no primary barrier?

Because a CT scanner never emits a useful beam toward a room surface. The fan or cone beam is collimated to the detector ring, rotates with the tube around the patient, and is absorbed by the patient and the gantry assembly. There is no fixed beam direction and no collimated beam leaving the gantry aperture, so there is nothing to place a primary barrier against. Every wall, floor, and ceiling of a CT room is a secondary barrier against scatter and leakage.¹

What is the difference between the DLP method and the isodose-map method?

The DLP method estimates scatter at a point from the weekly dose-length product multiplied by a scatter factor and corrected for distance; it is a single-point calculation that is simple and conservative. The isodose-map method uses the manufacturer's measured contours of scattered air kerma around the specific scanner, which capture the real spatial distribution including gantry and couch shadowing. NCRP 147 accepts both; the isodose method is generally the more spatially accurate, particularly behind and beyond the gantry.⁵⁷

How is required transmission converted into a material thickness?

The required transmission B is found from the design goal divided by the occupancy-weighted unshielded air kerma. B is then converted to thickness using the Archer three-parameter broad-beam model with the material-specific fitting parameters in NCRP 147, or, for a first estimate, by counting half-value layers as the base-2 logarithm of 1/B. The same B can be met with lead, concrete, or gypsum by reading each material's own transmission data.¹⁵

Does leakage radiation matter for CT shielding?

Leakage is included but is usually small compared with patient scatter at diagnostic CT techniques. The tube-housing leakage is bounded by the FDA performance standard, and the scattered radiation from the patient dominates the secondary-barrier requirement in nearly all directions. A complete analysis sums scatter and leakage, but the scatter term normally sets the thickness.¹²

How does a multi-detector scanner change the shielding versus an older single-slice unit?

Wider beams and higher output raise the scatter emitted per scan, roughly in proportion to beam width, so a modern multi-detector scanner can produce several times the scatter of an older single-slice machine at comparable technique. Shielding that was adequate for the old unit is not automatically adequate for its replacement, which is why a scanner swap triggers a fresh shielding review using the new unit's output and isodose data.⁹

What happens during the post-construction survey?

A qualified medical physicist operates the installed scanner under representative clinical technique and measures scattered air kerma at occupied locations on the far side of each barrier, paying particular attention to doors, viewing windows, conduit penetrations, and construction joints. The measured weekly air kerma, scaled by the projected workload and occupancy, is compared with the design goal. If any location exceeds the goal, additional shielding is added and re-surveyed before clinical use.¹³

Key Takeaways

A CT room is an entirely secondary-barrier installation: no useful beam escapes the gantry, so there is no primary barrier—only scatter and leakage to shield.¹
Workload is expressed as CT output (weekly DLP or mAs), not as a primary-beam tube loading directed at a wall.¹²
The DLP scatter method estimates unshielded air kerma as a scatter factor times weekly DLP, corrected by inverse-square distance; required transmission is the design goal divided by the occupancy-weighted result.¹⁶
NCRP 147 design goals are 0.1 mGy/week (controlled) and 0.02 mGy/week (uncontrolled), both below the 10 CFR Part 20 dose limits.¹⁴
Transmission is converted to thickness with the Archer three-parameter model or, approximately, by half-value layers; a busy body-CT wall against a full-occupancy neighbor often needs roughly 1–2 mm of lead.⁵
Manufacturer isodose maps capture the true spatial scatter distribution and are the preferred cross-check; the DLP estimate is conservative but can underestimate in some geometries.⁷⁸
CT is FDA- and state-regulated as an x-ray machine; a shielding plan review and a post-construction survey by a qualified medical physicist confirm the design.¹

Conclusion

CT room shielding is conceptually simpler than radiographic shielding in one respect—there is no primary barrier to place—but it demands a different and careful quantitative approach built on the scanner's output. The method is well established in NCRP Report No. 147: express the workload as weekly dose-length product, estimate scatter with the DLP scatter factor or the manufacturer's isodose maps, apply distance and occupancy, convert the required transmission into a material thickness with the Archer broad-beam model, and verify the result with a post-construction survey. Done correctly, the design protects the public and staff at a fraction of the cost of fixing an inadequate barrier after the walls are closed, and it stands up to the state plan review that a CT installation must pass.

How DRPS Can Help

DRPS designs and verifies CT room shielding end to end, from the workload assessment and barrier calculation through the stamped shielding plan to the post-construction survey that confirms it. Our board-certified medical physicists prepare documentation that meets state plan-review requirements and translates NCRP 147 methodology into a buildable specification. Explore our radiation shielding design, CT physics testing, and medical physicist consulting services. DRPS serves Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware—see our locations for details.

Related Resources

Diagnostic X-ray room shielding: NCRP 147 — the primary-plus-secondary counterpart for radiographic and fluoroscopic rooms.
PET/CT shielding calculations guide — combining 511 keV annihilation photons with CT scatter in one suite.
Lead shielding design principles — how lead attenuates diagnostic-energy photons.
Buildup factor and broad-beam gamma shielding — the broad-beam physics behind transmission curves.
Radiation protection shielding survey — verifying installed barriers after construction.
Radiation shielding design service — DRPS barrier design and plan review.

References

National Council on Radiation Protection and Measurements. NCRP Report No. 147: Structural Shielding Design for Medical Imaging X-Ray Facilities. 2004. (Supersedes Report No. 49.) ncrponline.org
Burrage JW, Causer DA. Comparison of scatter doses from a multislice and a single slice CT scanner. Australas Phys Eng Sci Med. 2006;29(3):257-259. doi:10.1007/BF03178574. doi.org
Archer BR, Thornby JI, Bushong SC. Diagnostic X-ray shielding design based on an empirical model of photon attenuation. Health Phys. 1983;44(5):507-517. doi:10.1097/00004032-198305000-00005. doi.org
U.S. Nuclear Regulatory Commission. 10 CFR Part 20: Standards for Protection Against Radiation. ecfr.gov
DeLorenzo MC, Yang K, Wu DH, Wang W, Rutel IB. Comparison of computed tomography shielding design methods using RadShield. J Radiol Prot. 2017;37(2):492-505. doi:10.1088/1361-6498/aa6c71. doi.org
Wallace H, Martin CJ, Sutton DG, Peet D, Williams JR. Establishment of scatter factors for use in shielding calculations and risk assessment for computed tomography facilities. J Radiol Prot. 2012;32(1):39-50. doi:10.1088/0952-4746/32/1/39. doi.org
Moreno C, Cenizo E, Bodineau C, Mateo B, Ortega EM. Analysis of shielding calculation methods for 16- and 64-slice computed tomography facilities. J Radiol Prot. 2010;30(3):557-565. doi:10.1088/0952-4746/30/3/010. doi.org
Watanabe H, Noto K, Shohji T, et al. A new shielding calculation method for X-ray computed tomography regarding scattered radiation. Radiol Phys Technol. 2016;10(2):213-226. doi:10.1007/s12194-016-0387-9. doi.org
Fog LS, Cormack J. Mathematical modeling of the radiation dose received from photons passing over and through shielding walls in a PET/CT suite. Health Phys. 2010;99(6):769-779. doi:10.1097/HP.0b013e3181e47a39. doi.org
U.S. Food and Drug Administration. 21 CFR 1020.33: Computed Tomography (CT) Equipment. ecfr.gov
National Council on Radiation Protection and Measurements. NCRP Report No. 49: Structural Shielding Design and Evaluation for Medical Use of X-Rays and Gamma Rays of Energies up to 10 MeV. 1976. (Superseded by Report No. 147.) ncrponline.org
National Institute of Standards and Technology. XCOM: Photon Cross Sections Database. nist.gov