Skip to main content
NCBI home page
RSNA Journals logoLink to RSNA Journals
. 2015 May 19;277(1):230–235. doi: 10.1148/radiol.2015142876

Marrow Adipose Tissue Quantification of the Lumbar Spine by Using Dual-Energy CT and Single-Voxel 1H MR Spectroscopy: A Feasibility Study

Miriam A Bredella 1,✉, Scott M Daley 1, Mannudeep K Kalra 1, J Keenan Brown 1, Karen K Miller 1, Martin Torriani 1
PMCID: PMC4613879  PMID: 25988401

Dual-energy CT can be used to assess marrow adipose tissue (MAT) content and bone mineral density (BMD) of the lumbar spine in a single examination; dual-energy CT can be used to correct for MAT, thereby providing more accurate assessment of BMD compared with single-energy quantitative CT.

Abstract

Purpose

To test the performance of dual-energy computed tomography (CT) in the assessment of marrow adipose tissue (MAT) content of the lumbar spine by using proton (hydrogen 1 [1H]) magnetic resonance (MR) spectroscopy as a reference standard and to determine the influence of MAT on the assessment of bone mineral density (BMD).

Materials and Methods

This study was institutional review board approved and complied with HIPAA guidelines. Written informed consent was obtained. Twelve obese osteopenic but otherwise healthy subjects (mean age ± standard deviation, 43 years ± 13) underwent 3-T 1H MR spectroscopy of the L2 vertebra by using a point-resolved spatially localized spectroscopy sequence without water suppression. The L2 vertebra was scanned with dual-energy CT (80 and 140 kV) by using a dual-source multi–detector row CT scanner with a calibration phantom. Mean basis material composition relative to the phantom was estimated in the L2 vertebra. Volumetric BMD was measured with and without correction for MAT. Bland-Altman 95% limits of agreement and Pearson correlation coefficients were calculated.

Results

There was excellent agreement between 1H MR spectroscopy and dual-energy CT, with a mean difference in fat fraction of −0.02 between the techniques, with a 95% confidence interval of −0.24, 0.20. There was a strong correlation between marrow fat fraction obtained with 1H MR spectroscopy and that obtained with dual-energy CT (r = 0.91, P < .001). The presence of MAT led to underestimation of BMD, and this bias increased with increasing MAT content (P < .001).

Conclusion

Dual-energy CT can be used to assess MAT content and BMD of the lumbar spine in a single examination and provides data that closely agree and correlate with 1H MR spectroscopy data.

© RSNA, 2015

Introduction

Studies have shown an important link between bone and fat. Both bone and fat cells arise from a common mesenchymal stem cell within bone marrow, which is capable of differentiating into osteoblasts and adipocytes (1,2). Many osteoporotic states, including old age, immobility, obesity, and anorexia nervosa, are associated with increased marrow adiposity (1,3,4), suggesting that marrow adipose tissue (MAT) may contribute to decreased bone strength and increased fracture risk. To gain further insights into the bone-fat connection, it would be desirable to study both bone mineral density (BMD) and MAT content in a single examination.

Furthermore, intravertebral MAT can affect the assessment of BMD with single-energy quantitative computed tomography (CT), a major determinant of fracture risk (5). Therefore, the assessment of MAT may become an important clinical tool to assess BMD accurately in states of increased marrow adiposity.

MAT can be quantified reliably by using single-voxel proton (hydrogen 1 [1H]) magnetic resonance (MR) spectroscopy (4,6,7); however, the evaluation of larger areas of heterogeneous marrow is limited. Advances in dual-energy CT allow the quantification of BMD of the lumbar spine (8) with no additional radiation dose compared with standard single-energy quantitative CT. The purpose of our study was to test the performance of dual-energy CT in the assessment of MAT content of the lumbar spine by using 1H MR spectroscopy as a reference standard and to determine the influence of MAT on the assessment of BMD by means of single-energy quantitative CT. We hypothesized that dual-energy CT, performed for BMD quantification, can also be used to quantify MAT of the lumbar spine accurately and that assessment of MAT would improve the accuracy of BMD evaluation with quantitative CT.

Materials and Methods

Our study was institutional review board approved and complied with Health Insurance Portability and Accountability Act guidelines. Written informed consent was obtained from all subjects after the nature of the procedure had been explained fully. Authors who were not employees of Mindways Software, Austin, Tex, had control of inclusion of any data.

Subjects

This was a prospective study. Subjects were part of an obesity trial and were recruited through advertisements from October 31, 2014, to June 9, 2014. Inclusion criteria were age of 18–65 years, body mass index of at least 25 kg/m2, and osteopenia (t score between −1 and −2.5 according to dual-energy x-ray absorptiometry). Exclusion criteria were history of osteoporosis, smoking, hypothalamic or pituitary disorders, diabetes mellitus, or other chronic illness; glucocorticoid use; or medications that are known to affect bone metabolism or BMD. After a screening visit in which subject eligibility was determined, all subjects underwent 1H MR spectroscopy, followed by dual-energy CT of the L2 vertebra. Studies were performed after an overnight fast to avoid potential influence of diet on MAT.

1H MR Spectroscopy of Bone Marrow

Subjects underwent 1H MR spectroscopy of the L2 vertebra to determine MAT content by using a 3.0-T MR imaging system (Trio; Siemens Medical Solutions, Malvern, Pa). Subjects were positioned feet first in the magnet bore, and a body matrix phased-array coil was positioned over the lumbar spine. A triplane localizer pulse sequence of the lumbar spine was performed with a repetition time (msec)/echo time (msec) of 15/5, with a section thickness of 3 mm. A voxel measuring 15 × 15 × 15 mm (3.4 mL) was placed within the anterior aspect of the L2 vertebral body; the basivertebral plexus was avoided. Single-voxel 1H MR spectroscopy data were acquired by using a point-resolved spatially localized spectroscopy pulse sequence without water suppression with the following parameters: 3000/30, eight acquisitions, 1024 data points, and receiver bandwidth of 2000 Hz. For each voxel placement, automated optimization of gradient shimming was performed.

Fitting of all 1H MR spectroscopy data was performed by using LCModel (version 6.3–0 K; Provencher, Oakville, Ontario, Canada). Data were transferred from the MR imaging unit to a Linux workstation, and metabolite quantification was performed by using eddy current correction and water scaling. A customized fitting algorithm for bone marrow analysis was used to provide estimates for all lipid peaks combined (0.9, 1.3, 1.6, 2.3, and 5.3 ppm). Lipid resonances were scaled to unsuppressed water peak (4.7 ppm) and expressed in MAT fraction.

Dual-Energy CT

All subjects underwent dual-energy CT of the L2 vertebra by using a second-generation dual-source 128-row multidetector CT scanner (Somatom Definition Flash; Siemens Medical Solutions, Forchheim, Germany). Subjects were positioned on a quantitative CTPro calibration phantom (Mindways Software). Helical scans were performed at 80 and 140 kV by using 210 and 80 mAs, respectively. Other scanning parameters included 1-second gantry rotation time, 0.9:1 pitch, and 64 × 0.6-mm detector configuration with double z-sampling. The images were reconstructed at 2-mm section thickness and 2-mm section interval by using the I31f reconstruction kernel with a sinogram-affirmed iterative reconstruction (SAFIRE; Siemens Healthcare) setting of 2. CT dose index volume, dose length product, and effective dose were 8 mGy, 32 mGy · cm, and 0.6 mSv, respectively.

Dual-energy CT measurements were used to derive basis material composition representation of the constituents of a measured volume. The mean CT value was assessed in spatially registered CT data sets acquired at 80 kVp and 140 kVp (CT80 and CT140 in the following equations) within an elliptical cylinder positioned in the anterior trabecular bone region of L2, which corresponded to the voxel placement from the 1H MR spectroscopy study. Basis material representation (ρH2O, ρK2HPO4) was then derived by using the following equations:

graphic file with name radiol.2015142876.eq1.jpg
graphic file with name radiol.2015142876.eq2.jpg

and

graphic file with name radiol.2015142876.eq3.jpg

where β terms are CT scanner–dependent responses for each of the basis materials determined from the CT calibration phantom measurements, C is the scanner-dependent offset, and D is a matrix determinant.

MAT fraction was determined by projecting the resulting basis material composition estimates onto a marrow-fat reference line defined relative to the BMD standard and yellow marrow and red marrow standards. The atomic composition of yellow marrow and red marrow was assumed to be that described in the International Commission on Radiation Units and Measurements Publication 46, with yellow marrow having an assumed physical density of 0.93 mg/cm3 and red marrow having an assumed physical density of 1.03 g/cm3 (9).

Single-energy volumetric BMD of trabecular bone was assessed in the same elliptical cylinder positioned in the anterior trabecular bone region of L2 by using the 140-kV data set. Three-dimensional reconstructive analysis was performed by using quantitative CT PRO software version 5.1 (Mindways Software). Additionally, marrow-corrected BMD was assessed by predicting BMD on the basis of replacing the observed marrow signal intensity with the expected signal intensity from an equivalent volume of water.

Statistical Analysis

Statistical analysis was performed by using MedCalc software (version 14; MedCalc, Mariakerke, Belgium). Variables were tested for normality of distribution by using the Shapiro-Wilk test. Results are reported as means ± standard deviations. Bland-Altman 95% limits of agreement for fat fraction obtained with 1H MR spectroscopy and dual-energy CT and those for BMD with and without removal of marrow elements were calculated. Linear correlation analysis was performed on the basis of the Pearson correlation coefficient to evaluate an association between fat fractions obtained with 1H MR spectroscopy and dual-energy CT and the difference in BMD at dual-energy CT with and without removal of marrow elements and fat fraction by using dual-energy CT and 1H MR spectroscopy. P values were derived for testing the hypothesis of no correlation against the alternative that there is a nonzero correlation. A difference with a P value of less than .05 indicated that the pairwise correlation was significantly different from zero.

Results

Our study group comprised 12 healthy subjects with a mean age of 43 years ± 13 (range, 24–60 years; median, 47 years) and a mean body mass index of 33 kg/m2 (range, 25–44 kg/m2). There were eight men (mean age, 41 years ± 13) and four women (mean age, 48 years ± 12).

Variables were normally distributed. There was no significant difference in MAT content measured with 1H MR spectroscopy and that measured with dual-energy CT. Mean MAT fraction with 1H MR spectroscopy was 0.70 ± 0.27, and mean MAT fraction with dual-energy CT was 0.71 ± 0.26 (P = .9). By using Bland-Altman analysis, there was excellent agreement between 1H MR spectroscopy and dual-energy CT without evidence of bias. The mean difference in fat fraction between the techniques was −0.02, with a 95% confidence interval; Pearson correlation coefficient was −0.24 to 0.20 (Fig 1).

Figure 1:

Figure 1:

Open in a new tab

Bland-Altman plot of MAT content with 1H MR spectroscopy (MRS) and dual-energy CT (DECT). All values are within limits of agreement (dotted lines), corresponding to ±1.96 standard deviation (SD) from the mean.

There was a strong positive correlation between marrow fat fraction obtained with 1H MR spectroscopy and dual-energy CT (r = 0.91, P < .001 [95% confidence interval: 0.70, 0.97]) (Fig 2).

Figure 2:

Figure 2:

Open in a new tab

Plot of correlation analysis of MAT content assessed with 1H MR spectroscopy (MRS) and dual-energy CT (DECT) . There is a strong correlation between the two techniques (r = 0.91, P < .001).

The single-energy data at 140 kV were used to assess volumetric BMD of L2, and the results were compared with BMD after removal of marrow elements as assessed with dual-energy CT. The presence of MAT led to underestimation of BMD, and this bias increased with increasing MAT content. Mean uncorrected BMD was 124.9 mg/cm3 ± 31.7, and mean corrected BMD after removal of marrow elements was 160.8 mg/cm3 ± 20.7. By using Bland-Altman analysis, the mean difference in corrected and uncorrected BMD was −35.9 mg/cm3 (−27.2%), with a 95% confidence interval of −83.4, 11.5.

There was a positive correlation between the difference in corrected and uncorrected BMD and MAT content assessed with dual-energy CT (r = 0.99, P < .001 [95% confidence interval: 0.97, 1.0]) (Fig 3a) and MAT content assessed with 1H MR spectroscopy (r = 0.87, P < .001 [95% confidence interval: 0.58, 0.96]) (Fig 3b), suggesting that increased MAT content leads to greater underestimation of BMD.

Figure 3a:

Figure 3a:

Open in a new tab

Plot of correlation analysis between the differences in corrected and uncorrected BMD and MAT content. (a) There is a strong correlation between the difference in corrected and uncorrected BMD and MAT content with dual-energy CT (DECT) (r = 0.99, P < .001). (b) There is a strong correlation between the difference in corrected and uncorrected BMD and MAT content with 1H MR spectroscopy (MRS) (r = 0.87, P < .001).

Figure 3b:

Figure 3b:

Open in a new tab

Plot of correlation analysis between the differences in corrected and uncorrected BMD and MAT content. (a) There is a strong correlation between the difference in corrected and uncorrected BMD and MAT content with dual-energy CT (DECT) (r = 0.99, P < .001). (b) There is a strong correlation between the difference in corrected and uncorrected BMD and MAT content with 1H MR spectroscopy (MRS) (r = 0.87, P < .001).

Discussion

Our study showed that dual-energy CT allows measurement of MAT content and BMD of the lumbar spine in a single examination. Dual-energy CT provides MAT data that closely agree and correlate with 1H MR spectroscopy data. Importantly, the presence of MAT leads to underestimation of BMD, and this bias increases with increasing MAT content. Dual-energy CT can be used to account for MAT content, thereby improving the accuracy of BMD assessment.

The noninvasive quantification of MAT has been proposed as a biomarker for stem cell differentiation into the bone and fat lineage and as a marker of skeletal integrity and fracture risk (1,7,10,11). In a study by Schellinger et al (12), patients with morphologic evidence of bone weakness, such as compression fractures or endplate depression, had 34% higher vertebral MAT content compared with control subjects. Therefore, an understanding of the factors that influence the development of MAT is of great clinical importance.

Dual-energy x-ray absorptiometry is most commonly used to assess BMD and fracture risk (13). However, BMD assessment with dual-energy x-ray absorptiometry is influenced by vascular calcifications, degenerative spine changes, or overlying soft tissues (14,15). In fact, dual-energy x-ray absorptiometry has been shown to be associated with a greater error in assessing BMD in obesity, owing to increased soft-tissue thickness compared with quantitative CT (16). Single-energy quantitative CT allows accurate assessment of true volumetric BMD of trabecular bone without the aforementioned limitations of dual-energy x-ray absorptiometry (17). However, the assessment of BMD with quantitative CT is affected by the presence of MAT (18,19). Dual-energy CT has been used in the past for assessment of MAT content (18–22), but there were concerns for increased radiation dose compared with single-energy quantitative CT (23).

Advances in CT technology allow quantification of BMD and MAT with no additional radiation dose compared with standard single-energy quantitative CT. In our study, the radiation dose was 0.6 mSv for the L2 vertebra, which is comparable to reported doses for standard single-energy quantitative CT (24). Reasons for reduced radiation dose in our study with dual-energy CT compared with previously reported doses with dual-energy CT include use of a more advanced CT scanner and iterative reconstruction technique (SAFIRE; Siemens Healthcare). The latter technique reduces image noise and thus allows dose reduction compared with other image reconstruction techniques (such as filtered back projection).

We found excellent agreement between dual-energy CT and 1H MR spectroscopy in the assessment of MAT content. An advantage of using 1H MR spectroscopy for assessment of MAT is the lack of ionizing radiation. However, a limitation of 1H MR spectroscopy is that it is confined to a single voxel and does not allow evaluation of larger heterogeneous marrow areas. Furthermore, 1H MR spectroscopy cannot be used for the quantification of BMD. Dual-energy CT allows assessment of both BMD and MAT.

The ability to quantify both MAT and BMD in a single examination can provide more information on the bone-fat connection and stem cell differentiation into the bone and fat lineage in normal and pathologic conditions. In addition, dual-energy CT could be used to monitor therapies that are known to influence the bone-fat connection, such as radiation therapy or chemotherapy (25).

Importantly, the quantification of MAT increased the accuracy of BMD compared with standard single-energy quantitative CT. In our study, the presence of MAT led to underestimation of BMD by 27%, which may have important clinical consequences, since patients may receive overdiagnoses of osteopenia or osteoporosis. The results from our study are similar to those in a study by Goodsitt et al, who found that single-energy quantitative CT leads to underestimation of BMD by about 23% compared with dual-energy CT (18). In a cadaveric study of vertebral bodies from 1988, Gluer et al found low MAT-related uncertainty for single-energy quantitative CT compared with dual-energy CT by using a GE CT/T 9800 scanner (GE, Milwaukee, Wis), suggesting that most studies for BMD in which this scanner is used can be performed by using single-energy CT (22).

Our study had several limitations. These include a small number of subjects with osteopenia at dual-energy x-ray absorptiometry and the inclusion of only the L2 vertebral body. Although our data were normally distributed, it could be that use of the Shapiro-Wilk test failed to reject the null hypothesis because of the small sample size. However, this was a feasibility study in which the performance of dual-energy CT was assessed in the quantification of MAT and BMD. Despite the low numbers, the data showed very strong agreement and correlations between the two methods. Further studies are necessary to test the performance of dual-energy CT in larger and heterogeneous populations and patients with osteoporosis. Strengths of our study are the detailed assessment of MAT with dual-energy CT and 1H MR spectroscopy and the detailed assessment of volumetric BMD with dual-energy CT and single-energy quantitative CT.

In conclusion, dual-energy CT can be used to assess MAT content and BMD of the lumbar spine in a single examination and provides data that closely agree and correlate with 1H MR spectroscopy data. The presence of MAT leads to underestimation of BMD, and this bias increases with increasing MAT content. Dual-energy CT can be used to correct for MAT, thereby providing more accurate assessment of BMD compared with single-energy quantitative CT.

Advances in Knowledge

  • ■ Dual-energy CT allows measurement of marrow adipose tissue (MAT) content of the lumbar spine and provides data that closely correlate with proton (hydrogen 1) MR spectroscopy data (r = 0.91, P < .001).

  • ■ Dual-energy CT can be used to assess both MAT and bone mineral density (BMD) in a single examination.

  • ■ Dual-energy CT can be used to correct for MAT content, thereby providing more accurate assessment of BMD compared with single-energy quantitative CT.

Implication for Patient Care

  • ■ The presence of MAT leads to underestimation of BMD, and this bias increases with increasing MAT content; dual-energy CT can be used to correct for MAT content, thereby providing more accurate assessment of BMD compared with single-energy quantitative CT.

Received January 9, 2015; revision requested February 6; revision received February 10; accepted February 25; final version accepted March 17.

From the 2014 RSNA Annual Meeting.

Funding: This research was supported by the National Institutes of Health (grant DK095792).

Disclosures of Conflicts of Interest: M.A.B. disclosed no relevant relationships. S.M.D. disclosed no relevant relationships. M.K.K. disclosed no relevant relationships. J.K.B. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: author is an employee, board member, and stockholder of Mindways Software. Other relationships: disclosed no relevant relationships. K.K.M. disclosed no relevant relationships. M.T. disclosed no relevant relationships.

Abbreviations:

BMD
bone mineral density
MAT
marrow adipose tissue

References

  • 1.Rosen CJ, Bouxsein ML. Mechanisms of disease: is osteoporosis the obesity of bone? Nat Clin Pract Rheumatol 2006;2(1):35–43. [DOI] [PubMed] [Google Scholar]
  • 2.Takada I, Suzawa M, Matsumoto K, Kato S. Suppression of PPAR transactivation switches cell fate of bone marrow stem cells from adipocytes into osteoblasts. Ann N Y Acad Sci 2007;1116:182–195. [DOI] [PubMed] [Google Scholar]
  • 3.Bredella MA, Fazeli PK, Miller KK, et al. Increased bone marrow fat in anorexia nervosa. J Clin Endocrinol Metab 2009;94(6):2129–2136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Bredella MA, Gill CM, Gerweck AV, et al. Ectopic and serum lipid levels are positively associated with bone marrow fat in obesity. Radiology 2013;269(2):534–541. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Laval-Jeantet AM, Roger B, Bouysee S, Bergot C, Mazess RB. Influence of vertebral fat content on quantitative CT density. Radiology 1986;159(2):463–466. [DOI] [PubMed] [Google Scholar]
  • 6.Li X, Kuo D, Schafer AL, et al. Quantification of vertebral bone marrow fat content using 3 Tesla MR spectroscopy: reproducibility, vertebral variation, and applications in osteoporosis. J Magn Reson Imaging 2011;33(4):974–979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Patsch JM, Li X, Baum T, et al. Bone marrow fat composition as a novel imaging biomarker in postmenopausal women with prevalent fragility fractures. J Bone Miner Res 2013;28(8):1721–1728. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wichmann JL, Booz C, Wesarg S, et al. Dual-energy CT-based phantomless in vivo three-dimensional bone mineral density assessment of the lumbar spine. Radiology 2014;271(3):778–784. [DOI] [PubMed] [Google Scholar]
  • 9.International Commission on Radiation Units and Measurements . Photon, electron, proton and neutron interaction data for body tissues. ICRU Report No. 46. Bethesda, Md: International Commission on Radiation Units and Measurements, 1992. [Google Scholar]
  • 10.Rosen CJ, Ackert-Bicknell C, Rodriguez JP, Pino AM. Marrow fat and the bone microenvironment: developmental, functional, and pathological implications. Crit Rev Eukaryot Gene Expr 2009;19(2):109–124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Schellinger D, Lin CS, Hatipoglu HG, Fertikh D. Potential value of vertebral proton MR spectroscopy in determining bone weakness. AJNR Am J Neuroradiol 2001;22(8):1620–1627. [PMC free article] [PubMed] [Google Scholar]
  • 12.Schellinger D, Lin CS, Lim J, Hatipoglu HG, Pezzullo JC, Singer AJ. Bone marrow fat and bone mineral density on proton MR spectroscopy and dual-energy x-ray absorptiometry: their ratio as a new indicator of bone weakening. AJR Am J Roentgenol 2004;183(6):1761–1765. [DOI] [PubMed] [Google Scholar]
  • 13.Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. WHO Study Group. Osteoporos Int 1994;4(6):368–381. [DOI] [PubMed] [Google Scholar]
  • 14.Bolotin HH. DXA in vivo BMD methodology: an erroneous and misleading research and clinical gauge of bone mineral status, bone fragility, and bone remodelling. Bone 2007;41(1):138–154. [DOI] [PubMed] [Google Scholar]
  • 15.Ho CP, Kim RW, Schaffler MB, Sartoris DJ. Accuracy of dual-energy radiographic absorptiometry of the lumbar spine: cadaver study. Radiology 1990;176(1):171–173. [DOI] [PubMed] [Google Scholar]
  • 16.Yu EW, Thomas BJ, Brown JK, Finkelstein JS. Simulated increases in body fat and errors in bone mineral density measurements by DXA and QCT. J Bone Miner Res 2012;27(1):119–124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Adams JE. Quantitative computed tomography. Eur J Radiol 2009;71(3):415–424. [DOI] [PubMed] [Google Scholar]
  • 18.Goodsitt MM, Johnson RH, Chesnut CH, 3rd. A new set of calibration standards for estimating the fat and mineral content of vertebrae via dual energy QCT. Bone Miner 1991;13(3):217–233. [DOI] [PubMed] [Google Scholar]
  • 19.Goodsitt MM, Rosenthal DI. Quantitative computed tomography scanning for measurement of bone and bone marrow fat content. A comparison of single- and dual-energy techniques using a solid synthetic phantom. Invest Radiol 1987;22(10):799–810. [DOI] [PubMed] [Google Scholar]
  • 20.Mayo-Smith W, Rosenthal DI, Goodsitt MM, Klibanski A. Intravertebral fat measurement with quantitative CT in patients with Cushing disease and anorexia nervosa. Radiology 1989;170(3 Pt 1):835–838. [DOI] [PubMed] [Google Scholar]
  • 21.Adams JE, Chen SZ, Adams PH, Isherwood I. Measurement of trabecular bone mineral by dual energy computed tomography. J Comput Assist Tomogr 1982;6(3):601–607. [DOI] [PubMed] [Google Scholar]
  • 22.Glüer CC, Reiser UJ, Davis CA, Rutt BK, Genant HK. Vertebral mineral determination by quantitative computed tomography (QCT): accuracy of single and dual energy measurements. J Comput Assist Tomogr 1988;12(2):242–258. [DOI] [PubMed] [Google Scholar]
  • 23.Marshall WH, Jr, Alvarez RE, Macovski A. Initial results with prereconstruction dual-energy computed tomography (PREDECT). Radiology 1981;140(2):421–430. [DOI] [PubMed] [Google Scholar]
  • 24.Damilakis J, Adams JE, Guglielmi G, Link TM. Radiation exposure in x-ray-based imaging techniques used in osteoporosis. Eur Radiol 2010;20(11):2707–2714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Carmona R, Pritz J, Bydder M, et al. Fat composition changes in bone marrow during chemotherapy and radiation therapy. Int J Radiat Oncol Biol Phys 2014;90(1):155–163. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Radiology are provided here courtesy of Radiological Society of North America

RESOURCES