What is quantitative CT (QCT)?

Quantitative CT (QCT) is a bone densitometry method that measures volumetric bone mineral density in milligrams per cubic centimeter from a CT scan acquired together with, or calibrated against, a reference phantom of known mineral density. It is most often applied to trabecular bone in the lumbar spine and, with three-dimensional acquisition, to the proximal femur.

Can the WHO T-score of -2.5 be used with spine QCT?

No. The World Health Organization definition of osteoporosis as a T-score of -2.5 or below was derived for DXA areal density and does not apply to volumetric spine QCT. Spine QCT uses absolute trabecular density thresholds instead. QCT of the hip can be processed to a DXA-equivalent areal density and T-score, which is one reason the hip is used when a T-score classification is required.

What density thresholds classify osteoporosis on spine QCT?

The American College of Radiology practice parameter uses absolute trabecular thresholds for the lumbar spine: a density above about 120 mg/cm3 is considered normal, 80 to 120 mg/cm3 indicates low bone mass, and below about 80 mg/cm3 is consistent with osteoporosis. These are density thresholds, not DXA T-scores.

What is opportunistic CT for bone density?

Opportunistic screening uses the CT attenuation of trabecular bone, typically measured in Hounsfield units in a vertebral body on a CT scan performed for another clinical reason, to flag possible low bone mass at no additional radiation dose. It is a screening signal, not a calibrated QCT measurement, and abnormal values should prompt dedicated bone density testing.

Does QCT require a calibration phantom?

Yes. QCT converts CT numbers to bone mineral density using a reference phantom of known densities, historically a solid dipotassium phosphate phantom. The phantom can be scanned simultaneously under the patient or asynchronously as part of a routine quality control program. Without traceable calibration, Hounsfield units alone are not a bone density measurement.

Who should set up and oversee a QCT program?

A QCT program runs on a CT scanner, so a qualified or board-certified medical physicist should establish the acquisition protocol, the calibration method, and the quality control schedule, and should verify reproducibility and cross-calibration. The CT scanner itself remains subject to state radiation-machine rules and, where applicable, ACR CT accreditation.

Quantitative CT (QCT) for Bone Mineral Density

Q: How is QCT different from DXA?

DXA produces an areal bone mineral density in grams per square centimeter from a projection image, so it is affected by bone size and overlying tissue. QCT produces a true volumetric density in milligrams per cubic centimeter from cross-sectional images, separates trabecular from cortical bone, and is not size dependent. QCT delivers a higher radiation dose and uses different diagnostic thresholds than DXA.

Quantitative CT (QCT) measures true volumetric bone mineral density in milligrams per cubic centimeter from a calibrated CT scan, separating trabecular from cortical bone in a way that projectional DXA cannot. Because the measurement is three-dimensional, it is not confounded by bone size or overlying tissue, but it carries a higher radiation dose than DXA and uses different diagnostic thresholds — so a defensible QCT program depends as much on calibration and quality control as on the scan itself.¹²

QCT has been in clinical use since the mid-1970s and remains a valuable densitometry tool where DXA is limited: large or degenerative spines, the need to resolve trabecular from cortical compartments, and body-composition research. This guide walks through the calibration physics, the QCT-versus-DXA tradeoffs, spine and hip measurement, opportunistic CT screening, radiation dose, regulatory context, and the quality control that keeps results reproducible.¹³

Introduction

Bone mineral density is a number, and the method that produces it determines what the number means. A DXA areal density in grams per square centimeter and a QCT volumetric density in milligrams per cubic centimeter are not interchangeable, and treating them as if they were is one of the most common errors in densitometry practice.¹²

Dual-energy X-ray absorptiometry (DXA) is the most widely used densitometry method and the basis for the World Health Organization T-score definition of osteoporosis. QCT answers a different question. It reconstructs cross-sectional images, isolates a trabecular-bone region of interest inside a vertebral body or the proximal femur, and converts the measured CT numbers to a mineral-equivalent density using a calibration phantom.¹²³

That difference is the whole story of QCT: better separation of bone compartments and freedom from size dependence, at the cost of higher dose, a dedicated calibration and analysis workflow, and diagnostic criteria that are not the familiar DXA T-score. This guide covers the physics, the clinical use, the practical optimization, and the regulatory framing that a facility needs before it reports a QCT density as a clinical result.

Topic Explanation

What is QCT?

QCT is the conversion of calibrated CT attenuation into bone mineral density. A CT scanner measures linear attenuation, reported as CT numbers in Hounsfield units (HU). Trabecular bone, cortical bone, marrow fat, and soft tissue all attenuate differently, and the attenuation of bone rises with its mineral content. If the scan is calibrated against a phantom whose rods contain known mineral-equivalent densities, the CT number inside a bone region of interest can be mapped to a bone mineral density (BMD) in milligrams per cubic centimeter.¹³

The most established application is the lumbar spine, where a region of interest is placed in the purely trabecular center of the vertebral body, typically across L1 to L3. Trabecular bone is metabolically active and responds earlier to disease and treatment than cortical bone, which is one reason spine QCT is sensitive to change. With three-dimensional multidetector acquisition, the same calibration logic extends to the proximal femur, where software projects the volumetric data into a DXA-equivalent areal density and T-score.¹³

Key terms used throughout this guide:

Volumetric BMD — density per unit volume (mg/cm3), the quantity QCT measures directly.
Areal BMD — density per unit projected area (g/cm2), the quantity DXA measures; it depends on bone size.
Trabecular region of interest — a volume placed inside the spongy center of a vertebral body to exclude cortical shell and posterior elements.
Calibration phantom — a reference object of known mineral-equivalent densities used to convert CT numbers to BMD.

Why volumetric density matters

A projection method such as DXA sums attenuation through the whole bone and divides by the projected area, so a physically larger bone returns a higher areal density even at the same true material density. QCT avoids this because it measures density within a defined volume. This makes QCT less sensitive to patient size, to vertebral body dimensions, and to overlying structures, and it allows the trabecular compartment to be measured on its own rather than averaged with the cortical shell and degenerative changes such as osteophytes and aortic calcification that can falsely elevate a DXA spine result.¹³

Key Technical Principles

From CT number to bone mineral density

The core of QCT is a linear calibration between the measured mean CT number in a region of interest and bone mineral density. Using a reference phantom whose rods span a range of known mineral-equivalent densities, a facility fits:

ρ_{BMD} = m \cdot \overline{HU}_{ROI} + c

where $ρ_{BMD}$ is the bone mineral density in mg/cm3, $\overline{HU}_{ROI}$ is the mean CT number in the trabecular region of interest, and $m$ (slope) and $c$ (intercept) are derived from the phantom rods. The slope and intercept are found from the phantom calibration line:

m = \frac{ρ _{2} - ρ _{1}}{HU _{2} - HU _{1}}

Worked spine QCT example

Consider an asynchronous calibration in which two phantom rods of known dipotassium phosphate (K2HPO4) equivalent density are measured:

Rod 1: $ρ_{1} = 0 mg/cm^{3}$ , measured mean $HU_{1} = 2$ .
Rod 2: $ρ_{2} = 200 mg/cm^{3}$ , measured mean $HU_{2} = 210$ .

The calibration slope is:

m = \frac{200 - 0}{210 - 2} = \frac{200}{208} \approx 0.962 mg/cm^{3} per HU

with intercept chosen so that the line passes through the rods, giving $c = - m \cdot HU_{1} \approx - 1.9 mg/cm^{3}$ . If the trabecular region of interest in a patient's L1 vertebral body has a mean CT number of 95 HU, the calibrated density is:

ρ_{BMD} = 0.962 \times 95 - 1.9 \approx 89 mg/cm^{3}

A trabecular spine density near 89 mg/cm3 falls in the low-bone-mass range by the absolute thresholds discussed below, which is a result that a projectional DXA of a large or degenerative spine might have missed. This example is simplified; a production QCT workflow uses a multi-rod calibration, consistent region-of-interest placement, and quality control of the scanner and phantom.¹³

Diagnostic thresholds and why the T-score changes

Spine QCT does not use the WHO DXA T-score. The WHO classification of osteoporosis as a T-score of -2.5 or below was derived from DXA areal density and does not transfer to volumetric trabecular density.²⁵ Instead, the American College of Radiology practice parameter applies absolute lumbar trabecular thresholds, summarized in the table below.⁴

QCT of the hip is the exception. Because three-dimensional femoral data can be projected into a DXA-equivalent areal BMD and compared with a reference population, hip QCT can be reported with a T-score and classified by the same -2.5 cut point as DXA, as set out in the 2015 ISCD Official Positions.³

QCT compared with DXA

Feature	Spine QCT	DXA
Quantity measured	Volumetric BMD (mg/cm3)	Areal BMD (g/cm2)
Bone compartment	Trabecular, isolated	Integral (cortical + trabecular)
Size dependence	Not size dependent	Larger bone reads higher
Degenerative artifact	Largely excluded from ROI	Osteophytes/calcification raise result
Diagnostic criterion	Absolute density thresholds	WHO T-score (-2.5)
Typical effective dose	Higher than DXA; comparable to a spine radiograph series	Well under 0.1 mSv
Calibration	Phantom, synchronous or asynchronous	Internal reference and daily QC

These tradeoffs follow directly from the physics: QCT's cross-sectional, phantom-calibrated measurement buys compartment separation and size independence, while DXA's low-dose projection buys speed, availability, and the established T-score framework.¹²³

Synchronous versus asynchronous calibration

In synchronous QCT, the calibration phantom lies under the patient during the scan, so the patient and phantom are imaged under identical conditions. In asynchronous QCT, the phantom is scanned separately — for example, as part of a weekly quality control routine — and the calibration is applied to patient scans acquired on the same scanner with the same protocol. Asynchronous calibration simplifies patient positioning and workflow but places a premium on CT number stability, which is why scanner quality control becomes part of the measurement rather than an afterthought.¹³

Clinical Impact

QCT is most valuable where DXA is least reliable. In patients with degenerative spine disease, scoliosis, prior instrumentation, large body habitus, or vascular calcification, a DXA spine result can be falsely elevated and clinically misleading. QCT's trabecular region of interest sidesteps those confounders and can reveal low bone mass that an areal measurement masks.¹³

Because trabecular bone turns over faster than cortical bone, spine QCT is also sensitive to change over time, which makes it useful for monitoring response to therapy when the measurement is reproducible. The three-dimensional data additionally support geometric and structural analysis and finite-element estimates of bone strength in research settings, information that an areal projection cannot provide.¹

The clinical cost is dose and standardization. A spine QCT delivers more dose than DXA, and because different scanners, protocols, and analysis packages can yield different absolute values, serial monitoring requires that the same scanner, protocol, and calibration be used each time. These are solvable problems, but they are physics and quality-control problems, which is why QCT belongs in a program with medical-physics oversight rather than being run ad hoc.¹³

Practical Optimization Tips

Standardize the acquisition

QCT reproducibility starts with a fixed protocol. Use a consistent tube potential (commonly 120 kVp for spine QCT so that the calibration and published thresholds apply), a fixed and documented tube-current and reconstruction setting, the same reconstruction kernel, and consistent patient positioning and centering. Changing kVp changes the relationship between CT number and mineral content and therefore the calibration, so protocol discipline is not optional.¹³

Place the region of interest consistently

The trabecular region of interest should sit in the central spongiosa of the vertebral body, clear of the cortical rim, the basivertebral vein, and any focal sclerosis or hemangioma. Consistent vertebral levels (for example, the mean of L1 and L2) and consistent region-of-interest size and placement are what keep serial measurements comparable. Automated and semi-automated analysis tools help, but they still require review.¹³

Treat calibration and QC as part of the measurement

Whether calibration is synchronous or asynchronous, the phantom and the scanner both need routine checks. Track the CT numbers of the phantom rods over time, watch for drift, and re-establish the calibration if the scanner is serviced, the tube is replaced, or the protocol changes. Precision should be characterized so that a least-significant-change can be reported for monitoring, exactly as it is for DXA.¹³

Use opportunistic CT as a signal, not a diagnosis

The trabecular attenuation of a vertebral body on a routine CT scan carries real information about bone status, and simple Hounsfield-unit thresholds in a vertebral body — for example, values in the broad range of roughly 110 to 160 HU at L1 — have been shown to flag low bone mass and osteoporosis opportunistically at no additional dose.⁶ This is a powerful case-finding tool, but an uncalibrated HU value is a screening signal, not a QCT measurement; abnormal values should lead to dedicated, calibrated bone densitometry.

Regulatory Considerations

QCT runs on a CT scanner, so it sits under radiation-machine regulation, not radioactive-material regulation. The CT system itself is a radiation-producing device regulated by state radiation-control programs, and the U.S. Food and Drug Administration regulates the equipment and the QCT analysis software as medical devices. There is no NRC byproduct-material involvement in QCT, because no radioactive material is used.

Several practical regulatory and accreditation points follow:

Equipment and accreditation. A CT scanner used for QCT is commonly enrolled in the ACR CT Accreditation Program and is subject to the facility's state radiation-machine rules, including acceptance testing and periodic medical-physicist surveys where required by the state.
Who performs the work. Clinical bone densitometry interpretation is a physician responsibility, while the physics — protocol design, calibration method, dose, reproducibility, and quality control — is the role of a qualified or board-certified medical physicist.
Standards of practice. The ACR-SPR-SSR practice parameter for musculoskeletal QCT and the ISCD Official Positions define the accepted methodology and reporting expectations for QCT of the spine and hip.³⁴
State context. DRPS supports facilities across Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware, where state radiation-control authorities administer the machine-registration, acceptance-testing, and survey requirements that apply to CT scanners used for QCT. Always confirm requirements with the authority having jurisdiction.

Frequently Asked Questions (FAQs)

Is QCT better than DXA?

Neither method is universally better; they answer different questions. DXA is lower dose, faster, more widely available, and underpins the WHO T-score framework, so it remains the default for screening. QCT measures true volumetric trabecular density and avoids size and degenerative confounders, so it is valuable where DXA is unreliable or where the trabecular compartment must be isolated.¹²³

Why does spine QCT not report a T-score?

The WHO T-score of -2.5 was derived from DXA areal density and does not transfer to volumetric trabecular density, so spine QCT reports absolute density thresholds instead. Hip QCT can be projected into a DXA-equivalent areal density and therefore can carry a T-score.²³⁵

What phantom does QCT use?

QCT historically uses a solid reference phantom containing dipotassium phosphate (K2HPO4) rods of known mineral-equivalent densities, scanned either simultaneously under the patient or asynchronously as part of quality control, to convert CT numbers to bone mineral density.¹³

How much radiation does spine QCT deliver?

Spine QCT delivers more dose than DXA, which is well under 0.1 mSv, but the dose compares favorably with other radiographic procedures such as a lumbar spine radiograph series performed in the same clinical setting. Peripheral QCT of the radius or tibia delivers a negligible dose.¹

Can a routine abdominal CT be used to check bone density?

Yes, as a screening signal. The trabecular attenuation of a vertebral body on a CT performed for another reason can flag low bone mass at no extra dose, but it is not a calibrated QCT result and abnormal values should prompt dedicated densitometry.⁶

Key Takeaways

QCT measures volumetric bone mineral density in mg/cm3 from a calibrated CT scan, isolating trabecular from cortical bone.¹³
Unlike DXA areal density, QCT is not size dependent and is largely free of degenerative-artifact confounding, at the cost of higher dose.¹²
The WHO T-score of -2.5 applies to DXA, not spine QCT; spine QCT uses absolute density thresholds, and hip QCT can report a DXA-equivalent T-score.²³⁴
Calibration — synchronous or asynchronous, against a K2HPO4-equivalent phantom — and scanner quality control are part of the measurement, not optional extras.¹³
Opportunistic CT attenuation is a valuable case-finding signal but is not a substitute for a calibrated QCT measurement.⁶
Because QCT uses a CT scanner, it falls under FDA and state radiation-machine regulation, with medical-physics oversight of protocol, calibration, dose, and QC.

Conclusion

Quantitative CT extends bone densitometry beyond what a projection method can do: a true volumetric, compartment-specific density that is not fooled by bone size or degenerative change. That capability is only as good as the calibration and quality control behind it, because the entire result rests on mapping CT numbers to mineral density on a stable, well-characterized scanner. Treated as a physics-governed program — fixed protocol, traceable calibration, consistent region-of-interest placement, and documented precision — QCT is a reliable and defensible tool for diagnosing and monitoring low bone mass where DXA falls short.¹³⁴

How DRPS Can Help

Diagnostic Radiation Physics Services (DRPS) supports imaging facilities across Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware with bone densitometry (DEXA) physics, CT physics testing, and medical physicist consulting — including QCT protocol design, calibration-phantom setup, reproducibility and cross-calibration studies, dose assessment, and the quality control that keeps serial measurements comparable.

A QCT program is a measurement system, not a single scan. DRPS helps facilities build the acquisition, calibration, and quality-control framework that makes QCT densities accurate, reproducible, and defensible over time.

Related Resources

References

Adams JE. Quantitative computed tomography. European Journal of Radiology. 2009;71(3):415-424. doi:10.1016/j.ejrad.2009.04.074. doi.org
Engelke K, Adams JE, Armbrecht G, et al. Clinical use of quantitative computed tomography and peripheral quantitative computed tomography in the management of osteoporosis in adults: the 2007 ISCD Official Positions. Journal of Clinical Densitometry. 2008;11(1):123-162. doi:10.1016/j.jocd.2007.12.010. doi.org
Engelke K, Lang T, Khosla S, et al. Clinical use of quantitative computed tomography (QCT) of the hip in the management of osteoporosis in adults: the 2015 ISCD Official Positions-Part I. Journal of Clinical Densitometry. 2015;18(3):338-358. doi:10.1016/j.jocd.2015.06.012. doi.org
American College of Radiology. ACR-SPR-SSR Practice Parameter for the Performance of Musculoskeletal Quantitative Computed Tomography (QCT). Revised 2018. Reston, VA: ACR. acr.org
World Health Organization. Assessment of Fracture Risk and Its Application to Screening for Postmenopausal Osteoporosis. WHO Technical Report Series 843. Geneva: WHO; 1994. who.int
Pickhardt PJ, Pooler BD, Lauder T, del Rio AM, Bruce RJ, Binkley N. Opportunistic screening for osteoporosis using abdominal computed tomography scans obtained for other indications. Annals of Internal Medicine. 2013;158(8):588-595. doi:10.7326/0003-4819-158-8-201304160-00003. doi.org
Adams JE, Engelke K, Zemel BS, Ward KA. Quantitative computer tomography in children and adolescents: the 2013 ISCD Pediatric Official Positions. Journal of Clinical Densitometry. 2014;17(2):258-274. doi:10.1016/j.jocd.2014.01.006. doi.org
International Society for Clinical Densitometry. 2015 ISCD Official Positions - Adult. Middletown, CT: ISCD. iscd.org
American College of Radiology. CT Accreditation Program Requirements. Reston, VA: ACR. acraccreditation.org
National Institute of Standards and Technology. X-Ray Mass Attenuation Coefficients and CT Number Traceability Resources. Gaithersburg, MD: NIST. nist.gov