AJR Get Involved! Great Benefits! Join ARRS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yoshizumi, T. T.
Right arrow Articles by Barnes, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yoshizumi, T. T.
Right arrow Articles by Barnes, L.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
DOI:10.2214/AJR.06.0742
AJR 2007; 188:1332-1336
© American Roentgen Ray Society

Original Research

Validation of Metal Oxide Semiconductor Field Effect Transistor Technology for Organ Dose Assessment During CT: Comparison with Thermoluminescent Dosimetry

Terry T. Yoshizumi1,2, Philip C. Goodman1, Donald P. Frush1, Giao Nguyen2, Greta Toncheva2, Maksudur Sarder3 and Lottie Barnes2

1 Department of Radiology, Duke University Medical Center, Box 3155, Durham, NC 27710.
2 Division of Radiation Safety, Duke University Medical Center, Durham, NC.
3 Radiation Safety Office, University of Arkansas, Fayetteville, AR.

Received June 2, 2006; accepted after revision December 6, 2006.

 
Address correspondence to T. T. Yoshizumi (yoshi003{at}mc.duke.edu).

FOR YOUR INFORMATION

The AJR has made getting the articles you really want really easy with a new online tool, Really Simple Syndication, available at www.ajronline.org. It's simple. Click the yellow RSS button located in the menu on the left of the page. You'll be on your way to syndicating your AJR content in no time.


Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
OBJECTIVE. The purposes of this study were to apply near-real-time dose-measurement technology with metal oxide semiconductor field effect transistors (MOSFETs) to the assessment of organ dose during CT and to validate the method in comparison with the thermoluminescent dosimeter (TLD) method.

MATERIALS AND METHODS. Dosimetry measurements were performed in two ways, one with TLDs and the other with MOSFETs. Twenty organ locations were selected in an adult anthropomorphic female phantom. High-sensitivity MOSFET dosimeters were used. For the reference standard, TLDs were placed in the same organ locations as the MOSFETs. Both MOSFET and TLD detectors were calibrated with an X-ray beam equivalent in quality to that of a commercial CT scanner (half-value layer, {approx} 7 mm Al at 120 kVp). Organ dose was determined with a scan protocol for pulmonary embolus studies on a 4-MDCT scanner.

RESULTS. Measurements for selected organ doses and the percentage difference for TLDs and MOSFETs, respectively, were as follows: thyroid (0.34 cGy, 0.31 cGy, -8%), middle lobe of lung (2.4 cGy, 3.0 cGy, +26%), bone marrow of thoracic spine (2.2 cGy, 2.5 cGy, +11%), stomach (1.0 cGy, 0.93 cGy, -6%), liver (2.5 cGy, 2.6 cGy, +6%), and left breast (3.0 cGy, 2.9 cGy, -1%). Bland-Altman analysis showed that the MOSFET results agreed with the TLD results (bias, 0.042).

CONCLUSION. We found good agreement between the results with the MOSFET and TLD methods. Near-real-time CT organ dose assessment not previously feasible with TLDs was achieved with MOSFETs. MOSFET technology can be used for protocol development in the rapidly changing MDCT scanner environment, in which organ dose data are extremely limited.

Keywords: CT dosimetry • metal oxide semiconductor field effect transistor • radiation dose • radiologic physics • thermoluminescent dosimeter


Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
There have been substantial advances in helical CT since its introduction in 1989 [1]. Today 4-, 8-, and 16-MDCT scanners are being replaced by 64-MDCT scanners. Organ dosimetry in MDCT, however, has lagged behind technologic advancement. Newer dosimetric technology entails use of metal oxide semiconductor field effect transistors (MOSFETs). Use of MOSFETs in radiation therapy dosimetry began in the middle 1990s. The initial applications of this new technology in diagnostic radiology, however, have been limited. MOSFETs were first used in diagnostic radiology in 1998, for patient skin dose measurement [2]. The use of MOSFET technology for CT dosimetry was proposed in 2003 by Yoshizumi et al. [3]. These authors discussed the following benefits of the use of MOSFETs in CT dosimetry: dose values are available immediately after scanning; repeated measurements are possible without delay; and the labor-intensive and time-consuming processes typically associated with use of thermoluminescent dosimeters (TLDs)—annealing, postirradiation wait, and review of individual chips—are eliminated [4-8]. Clinical applications of MOSFET technology in CT have begun to appear in peer-reviewed radiology journals [9-11], preceding publication of physics articles on the use of MOSFETs in CT dosimetry.

The purposes of this study were to apply MOSFET technology to CT organ dose assessment and to validate the MOSFET method in comparison with the TLD method as the standard of reference. To the best of our knowledge, this application has not been previously described.


Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
CT Scanner and Pulmonary Embolus Scan Protocol
Organ doses were measured on a 4-MDCT scanner (QX/i, GE Healthcare). We validated the measurements using the standard pulmonary embolus scan protocol (Table 1) at our institution.


View this table:
[in this window]
[in a new window]

  		TABLE 1 : Pulmonary Embolus Scan Protocol for 4-MDCT Scannera

 

Anthropomorphic Phantom
We used a commercially available anthropomorphic phantom for the study [12]. The specifications of the phantom were a woman 160 cm tall weighing 55 kg (Fig. 1). The phantom included bone, lung, and soft-tissue compositions [12]. It was subdivided into 38 contiguous sections each 2.5 cm thick and had attachable breasts and anatomic locations for 20 major organs. Each section contained 5-mm-diameter through holes, and hole locations were optimized for precise dosimetry of internal organs. Each hole in the phantom was labeled with a number referenced to in the manufacturer's user manual [12] for prevention of detector placement errors in the organs. Unused holes were filled with tissue-equivalent plugs. Each plug was easily cut into two pieces to hold the two TLD chips sandwiched between the two plugs. Each MOSFET detector was inserted in the desired organ hole, care being taken to protect fragile lead wires. For further protection, we placed a cushion of electrical tape under the wires (Fig. 2).


Figure 1
View larger version (57K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 1 —Photograph shows phantom (adult female phantom model 702-D, CIRS).

 

Figure 2
View larger version (122K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 2 —Photograph shows placement of metal oxide semiconductor field effect transistor detector in organ location.

 

Calibration of Dosimeters
The half-value layer for a single-detector CT scanner (CT/i, GE Healthcare) has been reported [13] to be 7.24 ± 0.02 mm Al at 120 kVp. We calibrated both MOSFETs and TLDs using a conventional radiographic X-ray tube. We matched the beam quality to the reported value [13] by adding 5-mm aluminum sheets to the tube (half-value layer {approx} 7 mm Al at 120 kVp). For calibration, our MOSFET and TLD dosimeters were placed side by side with an ion chamber (10x5-6 chamber, Radcal) and exposed simultaneously with a clinical CT beam equal to 140 kVp. Each exposure was measured with a monitor (MDH 1015, Radcal). Conversion from exposure (roentgens or millicoulombs per kilogram) to absorbed dose (rad or centigrays) was computed with the f-factor (0.94 cGy/R), chamber correction factor, and temperature and pressure correction factors.

MOSFET Dosimeters
For organ dose measurements, we placed high-sensitivity diagnostic MOSFET dosimeters (TN-1002RD, Best Medical Canada) in 20 organ locations. Doses were measured immediately after each scan, which can be considered near real-time conditions. For calibration, a group of MOSFET detectors were exposed three times at four exposure levels (50-1,500 mR [12.9-387 x 10-6 C/kg]), and the mean MOSFET readout (millivolts) and SD were computed. The least-squares fit routine (Prism, version 2.0, 1995, GraphPad Software) was used to obtain a calibration curve (MOSFET vs ion chamber measurement) by a fit of four MOSFET measurements to corresponding soft-tissue doses (centigrays) from an ion chamber. The slope of the linear regression curve gave a calibration factor for each MOSFET. Individual millivolt to centigray dose conversion factors (mV/cGy) for all 20 MOSFET detectors were computed with AutoSense software (Thomson-Nielsen TN-RD-49, Best Medical Canada) and stored in a laptop computer.

Uncertainty in MOSFET measurements was estimated on the basis of the measured percentage uncertainties (1 SD - 1{sigma}) as a function of MOSFET value. As shown in Figure 3, percentage SD increased as the dose to a MOSFET decreased. We selected the dose at the 25% SD level as the lower limit of detection (1.40 mGy) [1, 2]. The data were fitted to a one-phase exponential model as follows: y(%SD) = 98.78 x exp[-0.01472 x x(mrad)] + 11.08. Goodness of fit was R2 = 0.9790. MOSFET error bars were estimated with this equation.


Figure 3
View larger version (8K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 3 —Graph shows percentage uncertainties as function of metal oxide semiconductor field effect transistor (MOSFET) dose. Lower limit of detection is {approx} 140 mrad (1.4 mGy) at 25% at 1 SD. One-phase exponential fit: y(%SD) = 98.78 x exp[-0. 01472 x x(mrad) + 11.08.

 
TLDs
TLD-100 chips were used for the study. Our annealing procedure was as follows. TLDs were preheated for 1 hour at 400°C, then for 2 hours at 100°C. For TLD calibration, four sets of two TLD chips were exposed at four exposure levels (range, 50-1,500 mR [12.9-387 x 10-6 C/kg]), and two controls were used for background subtraction. The least-squares fit routine (Prism, version 2.0, Graph-Pad Software) was used to obtain a calibration curve (TLD vs ion chamber measurements) by fitting four TLD values (nanocoulombs) to corresponding softtissue doses (centigrays) from an ion chamber. TLD chips were assessed with an automatic TLD analyzer (Harshaw QS5500, Thermo Scientific) under nitrogen gas flow. All TLD measurements were obtained under nitrogen gas 24 hours after exposure. This step was used to avoid counting initial fading in the first 24 hours. For organ dose measurements, two TLD chips were placed within each organ location, and the mean of the two measurements was used for dose and uncertainty estimation (1 SD).

Statistical Analysis
Bland-Altman analysis was used to compare agreement between the two methods, that is, MOSFET versus TLD. All of the organ dose values in Table 2 were used to produce a Bland-Altman plot. The plot was a computation of the average of the two values (TLD and MOSFET for each organ) and the difference between the two measurements (Prism, version 4.0, 2005).


View this table:
[in this window]
[in a new window]

  		TABLE 2 : Comparison of CT Phantom Organ Dose Measurements Between Metal Oxide Semiconductor Field Effect Transistor (MOSFET) Technique and Thermoluminescent Dosimetry (TLD)

 


Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Table 2 shows the dose values obtained with TLDs and MOSFETs in all 20 organs. Measured organ dose values in the main CT scan region were as follows (expressed as TLD vs MOSFET and percentage difference): thyroid (0.34 vs 0.31 cGy, -8%), bone marrow in the thoracic spine (2.2 vs 2.5 cGy, +11%), bone marrow in rib location A (2.8 vs 2.6 cGy, -5%), bone marrow in rib location B (2.9 vs 2.9 cGy, -1%), left breast (3.0 vs 2.9 cGy, -1%), liver (2.5 vs 2.6 cGy, +6%), and stomach (1.0 vs 0.93 cGy, -6%). Figure 4 compares the TLD with the MOSFET absorbed dose measurements for each organ. Each error bar represents the uncertainty (1 SD) associated with the detection methods. Bland-Altman analysis revealed a bias of 0.042, and in 95% of organ doses, a difference between -0.913 and +0.997. Figure 5 shows the results of the Bland-Altman analysis. The bias is plotted as a dotted line at y = 0.042.


Figure 4
View larger version (11K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 4 —Graph shows absorbed doses for thermoluminescent dosimetry (white) versus metal oxide semiconductor field effect transistor (MOSFET) (gray) measurement. BM = bone marrow.

 

Figure 5
View larger version (5K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Fig. 5 —Graph shows Bland-Altman plot for difference between measurements versus averaged measurements. Dotted line shows bias (y = 0.042).

 

Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The Bland-Altman plot in Figure 5 displays the average of the two measurements (TLD and MOSFET) on the x-axis and the actual difference between the two measurements on the y-axis. The two methods had very similar results on average. The bias of nearly zero (0.042) indicates that the two methods are relatively comparable. Table 2 shows that organ dose values in the main CT scan region agreed well. We believe that the results show the feasibility and accuracy of MOSFET technology for CT radiation dose measurements, the percentage differences ranging from 1% to 27% in the field of view.

There was a factor-of-two difference in the left upper lobe of the lung (1.6 vs 0.88 cGy, -44%) and right breast (3.4 vs 1.8 cGy, 49%). We believe this difference is explained by the CT helical beam's not falling on the same organ location at the same point of the arc because of differences in the X-ray tube start angle between the TLD and MOSFET measurements.

The percentage differences increased for organs located away from the chest region. For example, bone marrow in the pelvis exhibited a 145% difference between TLD and MOSFET (0.013 vs 0.033 cGy). The large percentage difference was due to the difference in the lower limit of detection between TLD and MOSFET. We [3] have reported previously that the lower limit of detection for MOSFETs is approximately 1.40 mGy (140 mrad) at 25% (1 SD) compared with a TLD value of 0.10-0.20 mGy (10-20 mrad). In this example, the MOSFET measurement was 0.033 cGy (33 mrad), well below the lower limit of detection (140 mrad [0.14 cGy]). As a result, MOSFET technology has a limitation in low-dose applications. CT organ doses, however, are usually approximately 0.01-0.1 cGy. Therefore, this MOSFET limitation is not usually a problem for organ dose measurements within the scanned region.

We conclude that good agreement (bias, 0.042; Bland-Altman analysis) was found between MOSFET measurement and the TLD method; that the MOSFET technique allows near-real-time CT organ dose assessment, which was previously unfeasible; and that MOSFET technology is useful for new protocol development in the fast-changing field of MDCT technology.


Acknowledgments
 
We thank H. Mike Scribner, Radiation Oncology Physics Laboratory, for technical support with the TLD analyzer. We also thank Carolyn Lowry, Department of Radiology, for CT scan support.


References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Yoshizumi TT, Nelson RC. Radiation issues with multidetector row helical CT. Crit Rev Comput Tomogr 2003;44 : 95-117[CrossRef][Medline]
  2. Peet DJ, Pryor MD. Evaluation of a MOSFET radiation sensor for the measurement of entrance surface dose in diagnostic radiology. Br J Radiol 1999; 72:562 -568[Abstract]
  3. Yoshizumi TT, Sarder M, Goodman PC, Frush DP, Barnes L, Nguyen G. Application of MOSFET technology in CT organ dose assessment. (abstr) Med Phys 2003; 30:1422
  4. Paulson EK, Sheafor DH, Enterline DS, McAdams HP, Yoshizumi TT. CT fluoroscopy-guided interventional procedures: techniques and radiation dose to radiologists. Radiology 2001;220 : 161-167[Abstract/Free Full Text]
  5. Fricke BL, Donnelly LF, Frush DP, et al. In-plane bismuth breast shields for pediatric CT: effects on dose and imaging quality using experimental and clinical data. AJR 2003;180 : 407-411[Abstract/Free Full Text]
  6. Thornton FJ, Paulson EK, Yoshizumi TT, Frush DP, Nelson RC. Multidetector vs single detector CT: comparison of radiation dose and dose profile. Acad Radiol 2003;10 : 379-385[CrossRef][Medline]
  7. Gupta AK, Nelson RC, Johnson GA, Paulson EK, Delong DM, Yoshizumi TT. Optimization of eightelement multi-detector row helical CT technology for evaulation of the abdomen. Radiology2003; 227:739 -745[Abstract/Free Full Text]
  8. Heneghan JP, McGuire KA, Leder RA, DeLong DM, Yoshizumi T, Nelson RC. Helical CT for nephrolithiasis: comparison of the efficacy of conventional and reduced radiation dose techniques. Radiology2003; 229:575 -580[Abstract/Free Full Text]
  9. Hurwitz LM, Yoshizumi TT, Reiman RE, et al. Radiation exposure to the fetus from body MDCT during early gestation. AJR2006; 186:871 -876[Abstract/Free Full Text]
  10. Hurwitz LM, Yoshizumi TT, Reiman RE, et al. Radiation dose to the female breast from 16-MDCT body imaging protocols. AJR2006; 186:1718 -1722[Abstract/Free Full Text]
  11. Jaffe T, Nelson R, Johnson GA, et al. Optimization of multiplanar reformations from isotopic datasets acquired on a 16-slice multidetector helical CT scanner. Radiology 2006;238 : 292-299[Abstract/Free Full Text]
  12. Adult female phantom model 702-D handling instructions. Norfolk, VA: CIRS, 2002
  13. Kruger RL, McCollough CH, Zink FE. Measurement of half-value layer in x-ray CT: a comparison of two noninvasive techniques. Med Phys 2000; 27:1915 -1919[CrossRef][Medline]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Am. J. Roentgenol.Home page
T. A. Jaffe, T. T. Yoshizumi, G. I. Toncheva, G. Nguyen, L. M. Hurwitz, and R. C. Nelson
Early First-Trimester Fetal Radiation Dose Estimation in 16-MDCT Without and With Automated Tube Current Modulation
Am. J. Roentgenol., April 1, 2008; 190(4): 860 - 864.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
C. Coursey, D. P. Frush, T. Yoshizumi, G. Toncheva, G. Nguyen, and S. B. Greenberg
Pediatric Chest MDCT Using Tube Current Modulation: Effect on Radiation Dose with Breast Shielding
Am. J. Roentgenol., January 1, 2008; 190(1): W54 - W61.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Roentgenol.Home page
C. L. Hollingsworth, T. T. Yoshizumi, D. P. Frush, F. P. Chan, G. Toncheva, G. Nguyen, C. R. Lowry, and L. M. Hurwitz
Pediatric Cardiac-Gated CT Angiography: Assessment of Radiation Dose
Am. J. Roentgenol., July 1, 2007; 189(1): 12 - 18.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yoshizumi, T. T.
Right arrow Articles by Barnes, L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yoshizumi, T. T.
Right arrow Articles by Barnes, L.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS