ThompsonEtal2013.pdfAccurate localization of incidental fi.docx

ThompsonEtal2013.pdf
Accurate localization of incidental findings on the computed
tomography attenuation correction image: the influence
of tube current variation
John Thompsona,c, Peter Hogga, Samantha Highamb and David
Manninga,d
This observer performance study assessed lesion
detection in the computed tomography attenuation
correction image, as would be produced for myocardial
perfusion imaging over a tube current (mA) range. A static
anthropomorphic chest phantom containing simulated
pulmonary lesions was scanned using the four available
mA values (1, 1.5, 2 and 2.5) on a GE Infinia Hawkeye 4.
All other computed tomography acquisition parameters
remained constant throughout. Twenty-seven cases
showing zero to four lesions were produced for a
free-response receiver-operating characteristic method.
Image observations were completed using our novel

web-based ROCView software under controlled conditions.
The Jackknife alternative free-response receiver-operating
characteristic (JAFROC) figure of merit was used for
significance testing, wherein a difference in lesion
detection performance was considered significant
at P values less than 0.05. Twenty readers with varying
computed tomography experience (0–24 years) evaluated
108 images using an ordinal scale to score confidence.
The JAFROC analysis showed that there was no
statistically significant difference in performance between
mA values (P = 0.439) for this sample of observers.
In conclusion, no significant difference in lesion detection
performance was seen between the mA values. This
suggests that there is no value in using anything other
than the lowest mA value for the investigation of incidental
extracardiac findings. Nucl Med Commun 34:180–184 �c
2013 Wolters Kluwer Health | Lippincott Williams & Wilkins.
Nuclear Medicine Communications 2013, 34:180–184
Keywords: computed tomography acquisition parameters,

dose optimization, free-response receiver-operating
characteristic,
tube current
aUniversity of Salford, bPennine Acute Hospitals NHS Trust,
Greater
Manchester, cUniversity Hospitals of Morecambe Bay NHS
Foundation Trust,
Barrow-in-Furness and dLancaster University, Lancaster, UK
Correspondence to John Thompson, BSc (Hons), MSc, Nuclear
Medicine
Department, Furness General Hospital, Dalton Lane, Barrow-in-
Furness,
Cumbria LA14 4LF, UK
Tel: + 44 1229 870870 x54388; fax: + 44 1229 491036;
e-mail: [email protected]
Received 31 July 2012 Revised 28 September 2012
Accepted 29 October 2012
Introduction
Computed tomography (CT) has improved the sensitivity
and specificity of many nuclear medicine techniques
through the provision of additional anatomic information or
by providing a high-quality attenuation correction (AC)
map [1,2]. The use of AC is strongly recommended in some
patients undergoing certain procedures, most notably in
those undergoing myocardial perfusion imaging [3,4]. Within
this patient group there is also potential for the discovery of

extracardiac pathology by examining the coincidentally
produced computed tomography attenuation correction
(CTAC) image [4,5]. The ethical and legal considerations
of reviewing the CTAC image have been discussed
previously [6] and it is known that some nuclear medicine
centres routinely review these images to determine whether
incidental chest pathology is present. To a similar end, it has
also been suggested that the raw projection data from the
single-photon emission computed tomography (SPECT)
acquisition should be assessed for incidental cardiac and
extracardiac findings as part of the clinical routine [7,8].
Variation in tube current has no impact on tissue
attenuation values and Hounsfield units (HU) – the re-
sultant attenuation maps are therefore largely independent
of tube current (mA) [9–12]. However, the impact of
varying mA on lesion detection is less clear and this
deserves investigation because of the potential dose saving
at lower mA values. A suitable approach to investigating

this would be through visual performance assessment;
previous visual performance phantom simulations have
shown that lesion detection rates can be maintained at
reduced tube currents [13]. This paper describes a free-
response receiver-operating characteristic (FROC) investi-
gation of the full clinical mA range available on the GE
Infinia Hawkeye 4 SPECT/CT (GEH4) to establish the
impact on lesion detection performance within an anthro-
pomorphic chest phantom.
Materials and methods
Computed tomography attenuation correction
acquisitions/anthropomorphic phantom
The four mA values available for clinical use (1.0, 1.5, 2.0
and 2.5) on the GEH4 (General Electric Medical Systems,
Wisconsin, USA) were used to image a static anthropo-
morphic chest phantom, which contained simulated
pulmonary lesions [14] (Fig. 1). All other CT acquisition
parameters remained constant (Table 1).The lesions

simulated intrapulmonary pathology with diameters of 3,
5, 8, 10 and 12 mm at 630 – 800 and + 100 HU values. This
gave good representation of described density ranges for
Technical note
0143-3636 �c 2013 Wolters Kluwer Health | Lippincott
Williams & Wilkins DOI: 10.1097/MNM.0b013e32835c0984
Copyright © Lippincott Williams & Wilkins. Unauthorized
reproduction of this article is prohibited.
mailto:[email protected]
solid lesions (20–60 HU) [15–18] and ground-glass opacity
lesions (– 850 to – 450 HU) [19,20]. The phantom and
simulated lesion positions remained constant for the four
mA image acquisitions, ensuring the production of a case-
matched series of images suitable for FROC analysis.
A prestudy and poststudy diagnostic-quality CT scan was
acquired to ensure that no movement of simulated lesions
had occurred. These diagnostic-quality images also acted as
the FROC truth (gold standard/true lesion positions) to aid
accurate localization on the CTAC images. Observers were

blinded to these data.
Free-response receiver-operating characteristic
analysis
The FROC methodology is a significant improvement over
conventional receiver operating characteristic (ROC) tech-
niques. ROC investigations simply demand an observer to
determine whether an image contains lesions and assign
a confidence (rating) score to the image. FROC methods
allow observers to accurately localize multiple lesions
within a single image, with all localizations individually
scored. A proximity criterion, surrounding a lesion, is
applied to resolve ambiguities in lesion detection (lesion or
lesion mimic). This prevents nonlesion localizations (NL)
from being classified as successful lesion localizations
(LL) [21]; the methodology also allows multiple NL
marks to be made on an image. Image display and response
capture (IDRC) software, ROCView (Bury St Edmunds,
UK) [22], was applied in this observer performance study

to collect localization and confidence score (mark-rating
pairs) data. Data were analysed using a Jackknife
alternative free-response receiver-operating characteristic
(JAFROC) analysis [23] using the JAFROC figure of merit
(FOM) for optimal statistical power. A difference in lesion
detection performance would be considered significant at
P values less than 0.05.
Twenty observers with varying CT experience (0–24 years,
mean 4.25±6.78 years) performed the ROCview lesion
detection study. They assessed case-matched images [15
normal and 12 abnormal cases for each mA value (FROC
modality); 108 images in total] showing 17 simulated
pulmonary lesions. Observers were aware of the case mix
and the range of simulated pulmonary lesions per image
(0–4). Observers were able to make up to six localizations
per image, allowing opportunity for both LL and NL in all
images. Observers were instructed to locate only the
simulated pulmonary lesions and to ignore all other

coincidental mimics of pathology that the phantom may
produce.
Viewing conditions and study controls
The TG-18 test card (American Association of Physicists
in Medicine, College Park, Maryland, USA) [24,25] was
used to ensure the quality of the reporting standard
monitor used for displaying the images on ROCView. The
monitor was calibrated according to local clinical speci-
fications with ambient lighting dimmed and constant for
all observers. The influence of observer familiarity with
CT image adjustment (zooming/windowing) was negated,
as image adjustment was not permitted in this FROC
study. Consequently, the only variable was the mA value
used. ROCview automatically randomized and displayed
images on a CT lung window (width 1500, level – 500).
Dose recordings
The GEH4 provides the computed tomography dose
index (CTDIvol) as an indication of the dose received.

The effective mAs was calculated and the effective dose
(E) estimated using a chest-specific conversion factor
(0.014 mSv/mGy/cm) following a previously described
method [26].
Quality control – Hounsfield unit accuracy
Conventional CT indicates that HU variation is in the
magnitude of less than 1 HU for a variation in mA [9].
This was assessed on the GEH4 using the American
College of Radiologists CT Accreditation Phantom [27].
Fig. 1
A high-resolution computed tomography image of the phantom
showing
a + 100 HU simulated lesion (white arrow) in the right lung
field to
demonstrate the clinical value of the images used in this study.
Table 1 GE Infinia Hawkeye 4 SPECT/CT acquisition
parameters
(controls)
140 kVp,1.9 pitch, acquired slice thickness 4�5 mm, slice
reconstruction
6.1 mm, slice interval 3.4 mm, scan length 394.4 mm, scan FOV
565.65 mm,
display FOV 435 mm, pixel size 2.89 mm2, 2.0 rpm (L-mode),
matrix size

256�256
FOV, field of view; SPECT/CT, single-photon emission
computed tomography/
computed tomography.
Lesion localisation: influence of mA Thompson et al. 181
The average pixel value of a 200 mm
2
region of interest
was recorded for five modules of known density. HU
value accuracy is required for AC as they are converted to
attenuation coefficients at the energy of the SPECT
radionuclide [2].
Results
The JAFROC FOM revealed no significant difference in
lesion detection performance for any of the mA values
used (P = 0.826). Observer-averaged JAFROC FOM
values can be found in Table 2 alongside the dose
recordings and calculated E for each mA value. Individual
observer performance was consistent between variations

of mA with an intraobserver SD range of 0.01–0.11.
Interobserver variation was also low for each mA setting
(1 mA, SD = 0.08; 1.5 mA, SD = 0.09; 2.0 mA, SD = 0.10;
2.5 mA, SD = 0.13). E was calculated for full field of view
(40 cm) CTAC acquisitions at each mA setting, estimat-
ing a dose saving of 60% if using 1 mA instead of 2.5 mA.
CT HU values measured in a DICOM viewer [28]
showed negligible variation as a result of changing mA
(Table 3). The greatest deviation was observed within
the low-density regions of the phantom (Poly and Air),
although this was still small in magnitude (change of
3 HU). The variation in image quality at the four mA
values can be seen in Fig. 2.
Discussion
For the static anthropomorphic chest phantom represen-
tative of a man weighing 70 kg there was no statistically
significant difference in lesion detection for the four mA
values. Although our experiment did not simulate
respiratory motion, there is evidence to suggest that

incidental extracardiac lesions could be detected equally
well on images acquired at 1 mA as those acquired at
2.5 mA. This finding concurs with a previous patient-
based study, in which a 1 mA attenuation map was
acquired at rest and a 2.5 mA attenuation map was
acquired at stress; both image sets revealed abnormal
findings at a rate of 9.7% [12]. The patient-based study
and our work both describe potential dose saving in the
region of 60%.
We have shown, through the acquisition of QC data, that
HU accuracy is unaffected by mA in this SPECT/CT
system (Table 3); therefore, the linear attenuation
coefficients that make up the attenuation map must also
be unaffected. This is suggestive of the quality of AC
being maintained despite a reduced mA. Combining this
information with the results of our JAFROC analysis
suggests a potential for dose saving without any detri-
mental effect on the primary outcome (good-quality AC)

with respect to the CTAC acquisitions for this system or
secondary outcome (detection of incidental chest pathol-
ogy). The GEH4 offers image reconstruction filters that
are optimized for low-dose operation, and this system can
outperform diagnostic systems in terms of contrast-to-
noise ratio, in the low-dose range, because of the filters in
operation [29]. Consequently, it is possible that not all
systems will respond to using a very low-dose regime.
Our data provide insight into the potential for dose saving
in patients undergoing myocardial perfusion SPECT/CT.
However, some caution must be applied to the results
because our work only considers a phantom representing
‘Standard Man’; there is an acknowledged gap between
phantom studies and clinical work. In our (static) phantom
study we did not set out to determine the effect of
respiratory motion on lesion detection. Variation in
respiratory motion can be significant, both within subjects
and between them [30,31], but previous work comparing

2.5 mA attenuation maps at stress and 1 mA attenuation
maps at rest saw equal detection rates (9.7%) of abnormal
findings [12]. In this work the authors comment that the
attenuation map, with a rotation time of 14 s, was sampled
over approximately three respiratory cycles.
Respiratory motion has been a persistent problem in
hybrid imaging, affecting both emission and transmission
acquisitions with errors of misregistration known to
contribute to errors in AC [30,31]. Methods of correction,
including respiratory gating, deep inspiration breath-hold,
motion correction and postprocessing methods, have
been discussed to correct for these errors [30]. Further,
the phase of breathing (inspiration, midbreath, expira-
tions) also has been found to have a significant bearing on
AC [30,32].
Table 2 Dose recordings, calculations and free-response
receiver-operating characteristic results for all variations of mA
mA Effective mAs CTDI Estimated E (mSv) at 40 cm FOV
JAFROC FOM (95% CI)

1.0 15.789 1.587 0.89 0.568 (0.507–0.630)
1.5 23.684 2.380 1.33 0.590 (0.538–0.642)
2.0 31.579 3.173 1.78 0.585 (0.528–0.641)
2.5 39.474 3.967 2.22 0.591 (0.523–0.659)
CI, confidence interval; CTDI, computed tomography dose
index; FOM, figure of merit; FOV, field of view; JAFROC,
Jackknife alternative free-response receiver-operating
characteristic.
Table 3 Hounsfield unit accuracy at each mA value
Water Poly Acrylic Bone Air
True value (HU) 0 – 95 120 955 – 1000
mA
1.0 – 1 – 84 125 816 – 970
1.5 0 – 81 125 816 – 972
2.0 0 – 84 126 816 – 972
2.5 0 – 84 126 816 – 973
182 Nuclear Medicine Communications 2013, Vol 34 No 2
Further phantom work should attempt to simulate motion
over a range of respiratory amplitudes, with a previous
patient study suggesting that lesion position and size can
contribute to errors caused by respiratory motion [30].

Investigating the effect of patient size might prove
beneficial too.
Conclusion
Using a static anthropomorphic chest phantom with
simulated lesions, the GEH4 can be used with equal
confidence at each of its 4 mA settings for accurate LL on
the CTAC image. Before conducting a human study,
further phantom work should be carried out to consider
the effect of respiratory motion and patient size on lesion
detection.
Acknowledgements
The authors are grateful to Richard Lawson, Medical
Physicist at Central Manchester University Hospitals NHS
Foundation Trust, for clarification regarding CT acquisition
parameters of hybrid systems. The authors thank Katy
Szczepura, Medical Physicist and Senior Lecturer at The
University of Salford, for discussion around CT scanner
capability. The University of Cumbria kindly loaned the
Lungman phantom.

Conflicts of interest
There are no conflicts of interest.
References
1 Buck A, Nekolla S, Ziegler S, Beer A, Krause B, Herrmann K,
et al. SPECT/
CT. J Nucl Med 2008; 49:1305–1319.
2 O’Connor M, Kemp B. Single-photon emission computed
tomography/
computed tomography: basic instrumentation and innovations.
Semin Nucl
Med 2006; 36:258–266.
3 Heller G, Links J, Bateman T, Ziffer J, Ficaro E, Cohen M,
Hendel R.
American society of cardiology and society of nuclear medicine
joint
position: attenuation correction of myocardial perfusion SPECT
scintigraphy.
J Nucl Cardiol 2004; 11:229–230.
4 Flotats A, Knuuti J, Gutberlet M, Marcassa C, Bengel F,
Kaufmann P, et al.
Hybrid cardiac imaging: SPECT/CT and PET/CT. A joint
position statement
by the European Association of Nuclear Medicine (EANM), the
European
Society of Cardiac Radiology (ESCR) and the European Council
of Nuclear
Cardiology (ECNC) . Eur J Nucl Med Mol Imaging 2011;
38:201–212.

5 Goetze S, Pannu H, Wahl R. Clinically significant abnormal
findings on the
‘nondiagnostic’ CT portion of low-amperage-CT attenuation-
corrected
myocardial perfusion SPECT/CT studies. J Nucl Med 2006;
47:1312–1318.
Fig. 2
The anthropomorphic phantom scanned at each mA value,
showing low-density spherical simulated lesions (white arrows)
in the right lung (close
to vertebra) and left lung (posteriorly). Clockwise from top left:
1, 1.5, 2 and 2.5 mA.
Lesion localisation: influence of mA Thompson et al. 183
6 Tootell A, Vinjamuri S, Elias M, Hogg P. Clinical evaluation
of the computed
tomography attenuation correction map for myocardial
perfusion imaging:
the potential for incidental pathology detection. Nucl Med
Commun 2012;
33:1122–1126.
7 Chamarthy M, Travin M. Altered biodistribution and
incidental findings on
myocardial perfusion imaging. Semin Nucl Med 2010; 40:257–
270.
8 Gedik G, Ergün E, Aslam M, Caner B. Unusual extracardiac

findings
detected on myocardial perfusion single photon emission
computed
tomography studies with Tc-99m sestamibi. Clin Nucl Med
2007;
32:920–926.
9 Birnbaum B, Hindman N, Lee J, Babb J. Multi-detector row
CT attenuation
measurements: assessment of intra- and interscanner variability
with an
anthropomorphic body CT phantom. Radiology 2007; 242:109–
119.
10 Reza Ay M, Zaidi H. Computed tomography-based
attenuation correction in
neurological positron emission tomography: evaluation of the
effect of the
X-ray tube voltage on quantitative analysis. Nucl Med Commun
2006;
27:339–346.
11 Kamel E, Hany T, Burger C, Treyer V, Lonn A, von
Schulthess G, Buck A.
CT vs 68Ge attenuation correction in a combined PET/CT
system: evaluation
of the effect of lowering the CT tube current. Eur J Nucl Med
Mol Imaging
2002; 29:346–350.
12 Preuss R, Weiss R, Lindner O, Fricke E, Fricke H, Burchert
W. Optimisation
of protocol for low dose CT-derived attenuation correction in
myocardial
perfusion SPECT imaging. Eur J Nucl Med Mol Imaging 2008;
35:1133–1141.

13 Thompson J, Hogg P, Szczepura K, Manning D. Analysis of
CT acquisition
parameters suitable for use in SPECT/CT: a free-response
receiver
operating characteristic study. Radiography 2012; 18:238–243.
14 KyotoKagaku.com. Multipurpose Chest Phantom N1. Japan;
2011. Available
at: http://www.kyotokagaku.com/products/detail03/pdf/ph-
1_catalog.pdf
[Accessed 4 July 2012].
15 Das M, Mühlenbruch G, Katoh M, Bakai A, Salganicoff M,
Stanzel S, et al.
Automated volumetry of solid pulmonary nodules in a phantom:
accuracy
across different CT scanner technologies. Invest Radiol 2007;
42:297–302.
16 Ko J, Rusinek H, Jacobs E, Babb J, Betke M, McGuinness G,
Naidich D.
Small pulmonary nodules: volume measurement at chest CT –
phantom study. Radiology 2003; 228:864–870.
17 Marten K, Grabbe E. The challenge of the solitary pulmonary
nodule: diagnostic
assessment with multislice spiral CT. Clin Imaging 2003;
27:156–161.
18 Wormanns D, Kohl G, Klotz E, Marheine A, Beyer F,
Heindel W, Diederich S.
Volumetric measurements of pulmonary nodules at multi-row
detector CT:
in vivo reproducibility. Eur Radiol 2004; 14:86–92.

19 Oda S, Awai K, Murao K, Ozawa A, Yanaga Y, Kawanaka K,
Yamashita Y.
Computer-aided volumetry of pulmonary nodules exhibiting
ground-glass
opacity at MDCT. Am J Roentgenol 2010; 194:398–406.
20 Funama Y, Awai K, Liu D, Oda S, Yanaga Y, Nakaura T, et
al. Detection of
nodules showing ground-glass opacity in the lungs at low-dose
multidetector computed tomography: phantom and clinical
study. J Comput
Assist Tomogr 2009; 33:49–53.
21 Devchakraborty.com. Philadelphia; 2011. Available at:
http://
www.devchakraborty.com/Historical%20Background.html
[Accessed 4
July 2012].
22 Thompson J, Hogg P, Thompson S, Manning D, Szczepura K.
ROCView:
prototype software for data collection in Jackknife alternative
free-response
receiver operating characteristic analysis. Br J Radiol 2012;
85:1320–1326.
23 Devechakraborty.com. JAFROC 4.0.1. Philadelphia; 2012.
Available at:
http://www.devchakraborty.com/downloads [Accessed 4 July
2012].
24 Samei E, Badano A, Chakraborty D, Compton K, Cornelius
C, Corrigan K,
et al. Assessment of display performance for medical imaging
systems:
report of the American Association of Physicists in Medicine

(AAPM) Task
Group 18. AAPM On-Line Report No. 3. Madison, WI: Medical
Physics
Publishing; 2005.
25 Deckard.mc.duke.edu. Assessment of Display Performance
for Medical
Imaging Systems (American Association of Physicists in
Medicine (AAPM)
Task Group 18); 10 January 2010. Available at:
http://deckard.mc.duke.edu/
˜samei/tg18 [Accessed 31 January 2012].
26 Jessen K, Shrimpton P, Geleijns J, Panzer W, Tosi G.
Dosimetry for
optimisation of patient protection in computed tomography.
Appl Radiat Isot
1999; 50:165–172.
27 Gammex.com. CT Accreditation Phantom Instructions
(American College of
Radiologists). Wisconsin; 2011. Available at:
http://www.gammex.com/
acefiles/manuals/ACR_CT_Phantom.pdf [Accessed 11 July
2012].
28 Rosset A, Spadola L, Ratib O. OsiriX: an open-source
software for
navigating in multidimensional DICOM images. J Digit Imaging
2004;
17:205–216.
29 Hamann M, Aldridge M, Dickson J, Endozo R, Lozhkin K,
Hutton B.
Evaluation of a low-dose/slow-rotating SPECT-CT system. Phys
Med Biol

2008; 53:2495–2508.
30 Liu C, Pierce LA, Alessio AM, Kinahan PE. The impact of
respiratory motion
on tumor quantification and delineation in static PET/CT
imaging. Phys Med
Biol 2009; 54:7345–7362.
31 Dey J, Segars WP, Pretorius H, Walvick RP, Bruyant PP,
Dahlberg S, King
MA. Estimation and correction of cardiac respiratory motion in
SPECT in the
presence of limited-angle effects due to irregular respiration.
Med Phys
2010; 37:6453–6465.
32 Koshino K, Fukushima K, Fukumoto M, Sasaki K, Moriguchi
T, Hori Y, et al.
Attenuation correction for quantitative cardiac SPECT. Eur J
Nuc Med Mol
Imaging Res 2012; 2:33.
184 Nuclear Medicine Communications 2013, Vol 34 No 2
http://www.kyotokagaku.com/products/detail03/pdf/ph-
1_catalog.pdf
http://www.devchakraborty.com/Historical%20Background.html
http://www.devchakraborty.com/Historical%20Background.html
http://www.devchakraborty.com/downloads
http://deckard.mc.duke.edu/&tilde;samei/tg18
http://deckard.mc.duke.edu/&tilde;samei/tg18
http://www.gammex.com/acefiles/manuals/ACR_CT_Phantom.p
df

http://www.gammex.com/acefiles/manuals/ACR_CT_Phantom.p
df
PediatricDoseLimitationDuringCTscan.docx
DOSE LIMITATION FOR PEDIATRIC DURING CT SCAN
Dear
This is my professor feedback, So I want to edit proposal as
professor requests.
Using patients would be an insurmountable barrier to ethical
approval of the research.
We would like you to consider a phantom based study.
need to be more than a dose measuring exercise, you could look
at using exposure.
parameters that are outside the normal range (for example low
kV), or mA modulation
You could also consider image quality to see if the resulting
image is still of diagnostic quality.
Dose measurements/estimations could be used to establish the
risk of cancer induction.
We have attached several research articles published by our
Research group that will help you to re-develop your proposal.
And here my proposal
A research proposal
DOSE LIMITATION FOR PEDIATRIC DURING CT SCAN

Introduction
CT scan is an imaging modality which is being used widely in
the field of medicine. It is mostly employed for the diagnostic
purposes owing to the fact that it gives a detailed view of the
different structures of the body. It divides the body into
different cross sections, providing a detailed view of the
internal structures and organs. Apart from diagnostic purposes,
it is also being used for therapeutic purposes for therapeutic
taps and drainage of accumulated fluid etc. in the body cavities.
The basis of CT scan is the use of X-rays which form an image
of the body using the computerized program to contrast a three
dimensional image of the structures of the body based on the
passage of x-rays through them.
With the advances in the field of medicine and science, CT scan
has secured the position of being an important diagnostic
modality owing to the fact that it is a comparatively cheaper
investigation, yielding a detailed view of the internal structures
of the body, helping in the diagnosis and treatment of the
patient and having wide range of applications. It is thus the
most commonly advised investigation in cases where a
diagnosis is difficult to be established. It is being used in the
adult as well as the pediatric population. Due to the fact that the
CT scan employs use of computed X-rays to construct the three
dimensional images of the human body, the dose of the X-rays
is a growing concern. The radiations an individual is exposed to
during a single CT scan ranged from 10-20 mGy and can go up
to 80 mGy if a specialized CT image is required. This amount of
radiation is far greater than a routine X-ray. Exposing the
pediatric population to such high doses of radiations is the
concern behind this study.

Literature review
The basis of the CT scan and the method through which an
image of the various structures of the body are obtained is
through the use of X-rays which are used at different angles to
obtain a three dimensional image. If the dose of the radiation is
decreased to an absolutely a safe range for the pediatric
population, the purpose of the CT scan won’t be fulfilled. The
quality of the image would suffer tremendously and so the child
would be subjected to the radiations for no good outcome. Thus
a careful balance is required to be established between the
comparatively safe dose of radiation and the dose which is
required for optimal image production. This is how the dose
limitation can be achieved in children (Zacharias et al., 2013).
A study carried out to estimate the incidence of cancer among
the individuals who were exposed to the ionizing radiation
during a CT scan as children. They were assessed at later ages
and it was seem that the children who were exposed to increased
doses of radiations during CT scans, developed increased risk of
developing cancer later in life. The children who were exposed
to CT scans suffered from changes in their cells, leading to
mutations which caused cancers. These changes were more
dangerous among the pediatric age group because the cells were
dividing at a higher pace as compared to the cellular division in
older age groups (Miglioretti et al., 2013).
The dose which is used in the CT scans also varies among the
different CT machines. The latest equipment allows the
technician to adjust the dose of the radiation before exposing an
individual to unnecessary radiations. These devices are
extremely beneficial in adjusting the pediatric dose of radiation
according to the age of the patient because as the age of a child
increases, the harmful effects of the radiations decrease. But
this facility is not available in the old machines, most of which
are used in the community hospitals and imaging centers. These
facilities still use the same amount of radiations for every

individual which is contributing to the increase in the incidence
of increased pediatric cancers (Little, Grattan-Smith, &
Johnson, 2016).
The most beneficial method to decrease the exposure of high
dose of ionizing radiations during a CT scan in a child is by the
method called as “dose auditing”. This newly developed method
is based on the principle of audit and requires that each dose of
radiation which a person is exposed to should be audited against
the benefits of that exposure. The benefits which are achieved
from the exposure must outweigh the harmful effects of the dose
of radiation which the child is exposed to.
Aims/ hypothesis:
The aim of this study is to assess the safe reduced dose of the
radiation that a pediatric patient can be exposed to during a CT
scan. This safe dose should decrease the harmful effects of X-
rays exposure but still manage to provide the benefits of this
diagnostic modality. Another aim of this study is to assess the
detrimental effects which high dose radiations during CT scans
have already caused in the pediatric population.
Hypothesis:
To assess the reduction in the dose of radiation in pediatric
population and to assess the detrimental effects of high dose
radiation CT scan in pediatric population
Methods
The study will be carried out as a randomized controlled study.
The subjects selected for the study will be children from ages
between 1-5 and another group form 6-10. The study will be
carried out in two parts. The first part will require selection of
children from the two groups of ages as mentioned above whose
CT scan is planned. They will be divided in two groups,
randomly. The group A will have low dose of radiation
exposure and group B will have the standard pediatric dose
exposure. The results of the CT scan films will be compared.
As part two of the study, those people will be selected,
randomly form the community who have been exposed to the
radiations of CT scan as children. They will be categorized by

the age at which they were exposed to radiations and their graph
will also be plotted of the age at which they were exposed and
the dose of radiation which was given to them. These
individuals will also be questioned regarding any cancers they
suffered from. The data will give the idea of the risk of
development of cancer in individuals exposed to radiation as
children.
References
Duong, P. & Little, B. (2014). Dose Tracking and Dose
Auditing in a Comprehensive Computed Tomography Dose-
Reduction Program. Seminars In Ultrasound, CT And MRI,
35(4), 322-330. http://dx.doi.org/10.1053/j.sult.2014.05.004
Little, S., Grattan-Smith, D., & Johnson, B. (2016). CT
Radiation Dose Delivered by Community Hospitals and Imaging
Centers. Children's Healthcare Of Atlanta. Retrieved from
http://www.choa.org/Childrens-Hospital-
Services/Radiology/~/media/CHOA/Documents/Services/Radiol
ogy/Little-CT-dose-study.pdf
Miglioretti, D., Johnson, E., Williams, A., Greenlee, R.,
Weinmann, S., & Solberg, L. et al. (2013). The Use of
Computed Tomography in Pediatrics and the Associated
Radiation Exposure and Estimated Cancer Risk. JAMA
Pediatrics, 167(8), 700.
http://dx.doi.org/10.1001/jamapediatrics.2013.311
Zacharias, C., Alessio, A., Otto, R., Iyer, R., Philips, G.,
Swanson, J., &Thapa, M. (2013). Pediatric CT: Strategies to
Lower Radiation Dose. American Journal OfRoentgenology,
200(5), 950-956. http://dx.doi.org/10.2214/ajr.12.9026
ThompsonEtal2012bjr.pdf

SHORT COMMUNICATION
ROCView: prototype software for data collection in jackknife
alternative free-response receiver operating characteristic
analysis
1,2J THOMPSON, MSc, BSc(Hons), 2P HOGG, MPhil, FCR, 3S
THOMPSON, MEng, 2,4D MANNING, PhD, DSc and
2K SZCZEPURA, MSc, DipIPEM
1Department of Nuclear Medicine, Furness General Hospital,
Barrow-in-Furness, UK, 2Directorate of Radiography,
University of Salford, Salford, UK, 3Independent Software
Developer, and 4Division of Medicine, School of Health &
Medicine, Lancaster University, Lancaster, UK
ABSTRACT. ROCView has been developed as an image display
and response capture
(IDRC) solution to image display and consistent recording of
reader responses in
relation to the free-response receiver operating characteristic
paradigm. A web-based
solution to IDRC for observer response studies allows
observations to be completed
from any location, assuming that display performance and
viewing conditions are
consistent with the study being completed. The simplistic
functionality of the software
allows observations to be completed without supervision.
ROCView can display images

from multiple modalities, in a randomised order if required.
Following registration,
observers are prompted to begin their image evaluation. All data
are recorded via
mouse clicks, one to localise (mark) and one to score
confidence (rate) using either an
ordinal or continuous rating scale. Up to nine ‘‘mark-rating’’
pairs can be made per
image. Unmarked images are given a default score of zero.
Upon completion of the
study, both true-positive and false-positive reports can be
downloaded and adapted for
analysis. ROCView has the potential to be a useful tool in the
assessment of modality
performance difference for a range of imaging methods.
Received 11 September
2011
Revised 23 November 2011
Accepted 14 December
2011
DOI: 10.1259/bjr/99497945
’ 2012 The British Institute of
Radiology
The success of an imaging technique should not be judged
by physical measures, such as signal-to-noise ratio or contrast
resolution, alone. Human observations must be considered
because the reader is an integral part of the diagnostic
process. For the latter, receiver operating characteristic (ROC)
methods successfully quantify the combined performance of
imaging techniques and readers [1].

Traditional ROC analysis simply required readers to state
whether they thought a case was normal or abnormal, using
a rating scale to indicate decision confidence [2]. A notable
development in ROC methodology has been the free-
response ROC (FROC) method. Unlike ROC methods,
FROC uses location information to resolve ambiguities in
detection (lesion or lesion mimic) to prevent false identifica-
tion of pathology, resulting in a true-positive result [3].
Jackknife alternative free-response ROC (JAFROC)
methods [4] are the latest evolution of FROC methods
in which multiple readers and modalities are compared
with greater sensitivity to differences in performance
from other ROC methods [5]. Table 1 lists the terms
associated with JAFROC analysis. JAFROC methods
demand a precise response from the reader, in which
location and confidence information must be supplied
for each lesion within a case [6]. This identification and
scoring is typically referred to as a ‘‘mark-rating’’ pair [7].
Those setting up JAFROC studies are encouraged to
allow mouse clicks to record reader responses and an
acceptance radius around each area of pathology to
classify each response as either a lesion localisation/true
positive (LL/TP) or a non-lesion localisation/false
positive (NL/FP) [8].
JAFROC methods can be valuable in the evaluation of
different techniques in radiology. Like all observer
performance studies, achieving optimal statistical power
can require a large number of readers and cases [9, 10],
and a 50:50 ratio of normal and abnormal cases [11].
However, JAFROC methods achieve greater statistical
power than ROC for the same combination of readers
and cases [7]. This in turn requires a reliable method of

image display and response capture (IDRC). All ROC
methods require accurate data recording; ROCView is
proposed as a computer-based solution to accurate IDRC
in JAFROC analysis—the current end point of the free-
response paradigm.
Why develop ROCView?
A recent study of dose optimisation in CT [12]
prompted investigation into a reliable solution for
Address correspondence to: Mr John Thompson, Nuclear
Medicine
Department, Furness General Hospital, Dalton Lane, Barrow-in-
Furness LA14 4LF, UK. E-mail: [email protected]
The British Journal of Radiology, 85 (2012), 1320–1326
1320 The British Journal of Radiology, September 2012
IDRC suited to JAFROC methodology. Although other
software exists [13, 14], the authors took the opportunity
to develop ROCView to suit the needs of their current
research. As a web-based service ROCView would have
excellent availability with no software download or
installation required. A requirement of ROCView would
be to produce data suitable for analysis via JAFROC [15]
and Dorfman–Berbaum–Metz Multi-reader Multi-case
(DBM-MRMC) software [16] for direct comparison of the
performance of multiple readers and modalities.
Gathering suitable readers in a single centre for an
observer study also presented itself as a boundary to
conducting JAFROC research. Consequently, a geogra-
phically independent method of IDRC was required

to allow readers to complete the observation—this
prompted the development of a web-based solution.
ROCView meets the demands of geographical indepen-
dence and IDRC with the option to store resultant data in
a form suitable for DBM-MRMC analysis. The nature of
the software solution required a simplistic method to
maintain the integrity of the data, as there would be no on-
site support. Figure 1 describes the process of perception
that readers will use when assessing studies on ROCView.
Design and development of ROCView
ROCView is implemented as a web application, with
reader evaluations performed in a front-end display
system (browser/client), with a web server supplying
image data and recording reader responses into a
database. A second web interface provides administra-
tion facilities: user management, modality and case
management, creation of the ‘‘truth’’ and report genera-
tion. Guidance provided by Chakraborty [17] in relation
to conducting a JAFROC study aided the design of
ROCView. The development of a successful IDRC
system required an understanding of the following key
principles:
N image display
N comparing modalities
N setting the ‘‘truth’’
N acceptance radius criterion
N scoring confidence
N data recording.
The following description and the flow diagram
shown in Figure 2 explain the process for setting up
and running a JAFROC study on ROCView.

Table 1. The terminology associated with jackknife alternative
free-response receiver operating characteristic analysis
Term Alternative meaning/comment
Reader Observer, participant
Case Image
Lesion Pathology
Modality Variable
Case matched Same image displayed
Lesion localisation (LL) Successful identification of pathology
(TP)
Non-lesion localisation (NL) Unsuccessful identification of
pathology (FP)
Truth The true site of pathology in a case
Signal A stimulus within background noise stimulating a
localisation
Case memory The likelihood of a reader remembering the
appearance of cases shown in sequential order
Figure 1. The process used by readers attempting a study on
ROCView.
Short communication: ROCView: prototype software for data
collection in JAFROC analysis
The British Journal of Radiology, September 2012 1321
Image display
Image display is fairly straightforward for a JAFROC
study. Cases from multiple modalities are displayed, with
quick progression between them. ROCView automatically
randomises the images from all modalities in order to

reduce case memory. As a web-based application it is
important to ensure that the viewing conditions are
consistent and to a certain standard. A variety of monitor
test patterns can be used to ensure monitor performance
[18], in which it is also important to ensure a consistent
display response to allow a consistent response by the
observer. Methods to assess the adequacy of display
performance in accordance with the digital imaging and
communication in medicine (DICOM) greyscale standard
display function (GSDF) are available [19].
Comparing modalities
To allow direct comparison of two modalities, such as
a variation in image acquisition parameters, ROCView
allows the upload of a large number of image data sets
via file transfer protocol (FTP) [20]. All uploaded
modalities should be named or numbered appropriately
and all cases should be numbered and case-matched
across all modalities. When assigned to the appropriate
study, cases are automatically randomised. From this
point the study is ready for the ‘‘truth’’ to be set.
Setting the ‘‘truth’’
A suitably trained person should access the study as an
administrator and create the ‘‘truth’’, which is stored in a
database on ROCView to act as a standard with which all
further reader responses are compared, via the imple-
mentation of a pre-defined acceptance radius arising from
the pixel that is marked on the image. When performing
this task the administrator is able to view the modality
and case identifier alongside the image, together with the
localisation marker showing the acceptance radius, in
order to minimise error in lesion localisation. This is

particularly useful for phantom studies in which the
lesion sites are known. All readers are blinded to modality
and case identifiers. The administrator can make up to
nine lesion localisations per case.
Acceptance radius criterion
An acceptance radius is a region surrounding a lesion
that allows slight error in localisation. ROCView allows
adjustment of the acceptance radius in the source code. The
acceptance radius for current studies, determined by lesion
size, is nine pixels’ radial distance from the centre of the
localisation made by the administrator creating the ‘‘truth’’.
All marks within this radius are classified as LLs. All other
marks are NLs, penalising the reader for inaccuracies and
false localisations. This is, however, dependent on the
resolution of the monitor and the image size. Currently, the
largest size of image used has been 5126512 pixels, which
allows for a 1.8% error with an acceptance of 9 pixels in all
directions from the point of localisation. If monitor
resolution does not allow the maximum image size then
it will be scaled, with a 100% viewing window, and centred
over the cursor, available in the top right of the screen.
Figure 3 explains the implication of acceptance radius size;
a clinical example would equate to a vessel (lesion mimic)
next to a lesion in the thorax being incorrectly localised but
being classified as LL owing to it falling within the
acceptance radius of the lesion. Different values of radii
clearly affect reader performance [21], where a relaxed
criterion (20–40 pixels) improved apparent reader perfor-
mance. Localisation accuracy is an important distinction
between conventional ROC and free-response methods;
consequently, this criterion, at the discretion of the
researcher, must be controlled to ensure that JAFROC
maintains precision. When a reader clicks on the image in
ROCView a 19-pixel-diameter marker appears to aid the

reader’s judgement of accuracy (Figure 4). Existing phan-
tom-based ROCView studies have not used simulated
lesions that are .19 pixels in diameter. Therefore, localisa-
tion of the centre of the lesions would always result in a
TP/LL result.
Scoring confidence
Many variations of confidence scales are described
[22–24] and one must be mindful that one scale may not
Figure 2. The administration process used to create a study
on ROCView.
J Thompson, P Hogg, S Thompson et al
be suited to all investigations. To address this, ROCView
offers two types of scale: continuous and discrete
(Figure 5). When creating a study the administrator can
choose which type of scale will be presented to the
reader following a mark made on an image. The discrete
scale appears as a pop-up box showing a series of ranked
radio buttons (1–5, low to high confidence) from which
readers can make their selection. The continuous scale
presents as a slider bar that can function as a 101-point
scale similar to those in use [24]. The data can be
resampled to an 11-point discrete rating scale if there is
not a good distribution of responses—a problem con-
sistent with this type of scale arising from the inability to
estimate within a few percentage points [23].
Data recording

Once confidence has been scored, mark-rating pairs
are stored in a database (TP or FP) in the format shown
in Tables 2 and 3, which can then be downloaded and
analysed via JAFROC methods.
Figure 3. The implication of accep-
tance radius. A larger acceptance
radius could result in a greater
number of incorrect true-positive
results owing to the marking of a
lesion mimic that falls within the
acceptance radius. A small accep-
tance radius can reduce this risk and
would increase the number of false-
positive results owing to lesion
mimics and reader inaccuracies.
Figure 4. Screenshot of ROCView.
The main image (scaled) shows a
reader mark that prompted the
confidence scale to appear. Note
the option to remove unwanted
clicks. The image in the top right
of the screen shows the main image
at 100% size, without reader marks
for re-evaluation of uncertain areas.
Progress through the observation
challenge is also displayed.
Figure 5. Screenshot of ROCView showing the continuous-
style confidence scale variant. A mouse click prompts the
slider bar to appear. The slider starts at the left side of the
bar (score 1.0/low confidence) and can be moved to any
point along it to a maximum of 10.0 (high confidence).

For each successful localisation five integer values are
recorded in the TP sheet. Four integers are recorded for
each incorrect localisation in the FP sheet. The following
data need to be recorded as identifiable integers:
N reader ID
N case ID
N modality ID
N lesion ID (1, 2, 3 etc.)—not used in the FP worksheet
N confidence scale rating (TP or FP).
When no mark-rating pairs are made, a default score of
zero is stored for each case and lesion. These results are not
required for JAFROC analysis but have been included in
the reports to validate the results recorded by ROCView.
Currently the reader ID is recorded as the login details
(email address) of the reader. Once the data have been
downloaded, this can easily be converted to an integer in
Microsoft ExcelH (Microsoft Excel, Redmond, WA). Once
downloaded, TP and FP results need correlating with a
‘‘truth’’ sheet in a single Excel file as described by
Chakraborty and Yoon [25]. This data file can then be
analysed using JAFROC v. 4.0 [15] to transform the
FROC data to inferred ROC data (*.lrc file), which can be
analysed using DBM-MRMC software [16].
Software development

When the application starts, it makes an asynchronous
request to the web server to retrieve the modalities to be
evaluated by the user. This is dictated by the activation
key used when registering as a user, since ROCView
runs multiple studies concurrently. Cases are displayed
in turn, and, as the user creates mark–rating pairs, each
response is sent asynchronously to the server and is
stored immediately in the appropriate database (TP or
FP). The software has been made functional for general
purpose personal computers, in terms of the processing
power and memory required, as a result of the tools used
to develop the software (Table 4). As with all FROC
studies, monitor performance and viewing conditions
remain a concern and these should meet the required
standards consistent with the study objectives.
The client has been designed to use the native HTML
and JavaScript capabilities of browsers. Consequently,
browser plugins are not necessary and the hardware
requirements of the client remain low. The native solution
is at risk of manipulation by the user; for example, by
manipulation of the page uniform resource locators or by
reloading the evaluation page in unsupervised studies.
Therefore, a function is applied to the evaluation data to
maintain integrity between server requests. This prevents
the user from reloading the evaluation page and returning
to evaluate previous images while ensuring that evalua-
tions resume at the last stored point if they are not
completed in one attempt (due to loss of internet
connection or computer hardware failure).
Future application, flexibility and
development
ROCView has the potential to deal with any images

acquired in a manner suitable for FROC analysis. With
the capability to display images from a wide range of
acquisition methods (CT, ultrasound, MRI, nuclear
medicine, mammography and computed/digital radio-
graphy) there is a relatively unlimited application for
lesion detection-based studies. A phantom-based lesion
detection study of CT acquisition parameters highlighted
the type of research study for which this software is
suitable [12]. Producing case-matched images is a
significant challenge for MRMC studies, but if this can
be achieved ROCView can be a powerful tool for the
display of images and the accurate recording of reader
responses.
A significant development to the software would see
the creation of an acceptance volume, to allow classifica-
tion of LLs and NLs, for lesions that cross multiple cross-
sectional images. Software to allow this is available [14]
and has been used in a FROC study of mammography
techniques [34]. A further development would allow
hybrid images to be displayed in their component parts
Table 3. False-positive database format as recorded by
ROCView
Reader ID Modality ID Case ID Reader mark (x, y) FP rating
Name Variable Value Value Confidence
Example
[email protected] Low resolution 39 236 129 3
FP, false positive; ID, identity.
HTML format, all fields; CSV format, exclude italic fields.
Table 2. True-positive database format as recorded by
ROCView

Reader ID Modality ID Case ID Lesion ID Reader mark (x, y)
Truth (x, y)
Proximity
(pixels) TP rating
Name Variable Value Value Value Value Value Confidence
Example
[email protected] High resolution 4 1 121 293 120 294 1.4 4
ID, identity; TP, true positive.
HTML format, all fields; CSV format, exclude italic fields.
(emission, transmission and fused) to allow novel FROC
studies of diagnostic performance in single photon
emission CT and positron emission tomography/CT.
There is also scope for the development of a region-of-
interest style acceptance criterion to enable a form of
JAFROC analysis suited to fracture detection in which
the over-riding outcome would be a study in a similar
vein to mammography PERFORMS [35] to quantify
trainee performance in fracture detection. ROCView, as a
web-based service, requires no download of software.
ROCView is currently being used in a research pro-
gramme but may become available commercially to
other researchers in the future.
References

1. Gur D, Rockette H, Bandos A. ‘Binary’ and ‘non-binary’
detection tasks: are current performance measures optimal?
Acad Radiol 2007;14:871–6.
2. Manning D. Evaluation of diagnostic performance in
radiography. Radiography 1998;4:49–60.
3. devchakraborty.com [homepage on the internet].
Philadelphia: Dev Chakraborty; 2011 [updated 2011; cited
4 November 2011]. Available from: http://www.
devchakraborty.com/Historical%20Background.html
4. Chakraborty D. Analysis of location specific observer perfor-
mance data: validated extensions of the jackknife free-
response (JAFROC) method. Acad Radiol 2006;13:1187–93.
5. devchakraborty.com [homepage on the internet]. Phila-
delphia: Dev Chakraborty and Hong-Jun Yoon; 2011.
JAFROC 4.0 user manual [updated 2010; cited 10 August
2011]. Available from: http://www.devchakraborty.com/
Receiver%20operating%20characteristic.pdf
6. Zarb F, Rainford L, McEntree M. Image quality assessment
tools for optimisation of CT images. Radiography 2010;16:
147–53.
7. Chakraborty D. Validation and statistical power compar-
ison of methods for analysing free-response observer
performance studies. Acad Radiol 2008;15:1554–66.
8. Chakraborty D, Hong-Jun Y, Mello-Thomas C. Spatial
localization accuracy of radiologists in free-response stu-
dies: inferring perceptual FROC curves from mark–rating
data. Acad Radiol 2007;14:4–18.
9. Obuchowski N. Reducing the number of reader interpreta-

tions in MRMC studies. Acad Radiol 2009;16:209–17.
10. Chakraborty D. How many readers and cases does one
need to conduct an ROC study? Acad Radiol 2011;18:127–8.
11. Manning D, Leach J, Bunting S. A comparison of expert and
novice performance in the detection of simulated pulmon-
ary nodules. Radiography 2000;6:111–16.
12. Thompson J, Hogg P, Szczepura K, Tootell A, Sil J,
Manning
D. Determination of optimal CT exposure factors for lung
lesions using an anthropomorphic chest phantom for SPECT-
CT. Eur J Nucl Med Mol Imaging 2010;37(Suppl. 2):S494.
13. Jacobs F, Zanca R, Oyen H, Bosmans J. Sara, an advanced
software platform for human observer performance ex-
periments. In: Proceedings of the XIVth Medical Image
Perception Conference; 9–12 August 2011; Dublin, Ireland.
Available from: http://home.comcast.net/,eakmips/
update.htm (software available from: http://www.
kuleuven.be/radiology/lucmfr/sara/index.html)
14. Håkansson M, Svensson S, Zachrisson S, Svalkvist A, Båth
M, Månsson L. ViewDEX: an efficient and easy-to-use
software for observer performance studies. Radiat Prot
Dosim 2010;139:42–51. Software available from: http://
helios.ifss.gu.se/viewdex/index_v3_slide0001.htm
Philadelphia: Dev Chakraborty; 2011. JAFROC 4.0.1 analy-
sis software [updated 2011; cited 10 August 2011]. Available
from http://www.devchakraborty.com/downloads.html
16. perception.radiology.uiowa.edu [homepage on the internet].

Iowa City, IA: The Medical Image Perception Laboratory;
2011. DBM-MRMC v. 2.3 [updated 2011; cited 10 August
2011]. Available from: http://perception.radiology.uiowa.
edu/Software/ReceiverOperatingCharacteristicROC/
DBMMRMC/tabid/116/Default.aspx
Philadelphia:
Dev Chakraborty; 2011. How to conduct a free-response study
[updated 2011; cited 10 August 2011]. Available from: http://
www.devchakraborty.com/HowToConduct.html
18. Deckard.mc.duke.edu [homepage on the internet]. Durham,
NC: American Association of Physicists in Medicine; 2011.
Online report no. 3: assessment of display performance for
medical imaging systems [updated 2005; cited 10 August
2011]. Available from: http://deckard.mc.duke.edu/
,samei/tg18_files/tg18.pdf
19. Fetterly K, Blume H, Flynn M, Samei E. Introduction to
grayscale calibration and related aspects of medical
imaging grade liquid crystal displays. J Digit Imaging
2008;21:193–207.
20. filezilla-project.org [homepage on the internet]. Aachen,
Germany. Filezilla; 2011. Filezilla 3.5.0: The free FTP
solution [updated 2011 May 22; cited 10 August 2011].
Available from: http://filezilla-project.org/
21. Gur D, Bandos A, Klym A, Cohen C, Hakim C, Hardesty L,
et al. Agreement of the order of overall performance levels
under
different reading paradigms. Acad Radiol 2008;15:1567–73.
22. Park S, Goo J, Jo C-H. Receiver operator characteristic
(ROC) curve: practical review for radiologists. Korean J

Radiol 2004;5:11–18.
23. Hadjiiski L, Chan H-P, Sahiner B, Helvie M, Roubidoux M.
Quasi-continuous and discrete confidence rating scales for
observer performance studies: effects on ROC analysis.
Acad Radiol 2007;14:38–48.
24. Rockette H, Gur D. Selection of a rating scale in receiver
operating characteristic studies: some remaining issues.
Acad Radiol 2008;15:245–8.
Philadelphia: Dev Chakraborty; 2011. JAFROC 4.0 user
manual [updated 2010; cited 10 August 2011]. Available
from: http://www.devchakraborty.com/downloads.html
26. apache.org [homepage on the internet]. Los Angeles, CA:
Apache Software Foundation; 2011. Apache HTTP server
[updated 22 May 2011; cited 10 August 2011]. Available
from: http://httpd.apache.org/
Table 4. Programming tools and applications used in the
development of ROCView (open source components)
Tool Function
Apache HTTP Server [26] + PHP module [27] Web server
MySQL [28] Server database
Flex 4 SDK [29] + jQuery [30] Client applications, JavaScript
library
Apache Ant [31] Source file deployment
JSLint [32] Code quality checking of JavaScript
JSMin [33] Compression of client-side JavaScript code

27. php.net [homepage on the internet]. Denmark: PHP; 2011.
PHP Scripting Language [updated 28 June 2011; cited 10
August 2011]. Available from: http://www.php.net/
28. mysql.com [homepage on the internet]. Redwood Shores,
CA: MySQL; 2011. Open source database [updated 25 July
2011; cited 10 August 2011]. Available from: http://www.
mysql.com/
29. adobe.com [homepage on the internet]. San Jose, CA:
Adobe;
2011. Flex 4 SDK [updated 20 June 2011; cited 10 August
2011]. Available from: http://opensource.adobe.com/wiki/
display/flexsdk/Flex+SDK
30. jquery.com [homepage on the internet]. Boston, MA:
jQuery; 2011. JavaScript Library [updated 23 March 2010;
cited 10 August 2011]. Available from: http://jquery.
com/
31. ant.apache.org [homepage on the internet]. Los Angeles,
CA: Apache; 2011. Apache Ant Java Library [updated 27
December 2010, cited 10 August 2011]. Available from:
http://ant.apache.org/
32. jslint.com [homepage on the internet]. USA: JSLint; 2011.
JavaScript Code Quality Tool [updated 26 August 2011, cited
10 August 2011]. Available from: http://www.jslint.com/
33. crockford.com [homepage on the internet]. USA: JSMin;

2011.
JavaScript Minifier [updated 2003 Dec 04, cited 10 August
2011]. Available from: http://www.crockford.com/javascript/
jsmin.html
34. Svahn T, Andersson I, Chakraborty D, Svensson S, Ikeda D,
Förnvik D, et al. The diagnostic accuracy of dual-view digital
mammography, single-view breast tomosynthesis and a
dual-view combination of breast tomosynthesis and digital
mammography in a free-response observer performance
study. Radiat Prot Dosim 2010;139:113–17.
35. Scott H, Gale A. Breast screening: PERFORMS identifies
key
mammographic training needs. Br J Radiol 2006;79:S127–33.
TootellEtal2014Radiography.pdf
lable at ScienceDirect
Radiography 20 (2014) 323e332
Contents lists avai
Radiography
journal homepage: www.elsevier.com/locate/radi
An overview of measuring and modelling dose and risk from
ionising
radiation for medical exposures
Andrew Tootell*, Katy Szczepura, Peter Hogg

Directorate of Radiography, Salford University, Salford, M6
6PU, UK
a r t i c l e i n f o
Article history:
Received 24 February 2014
Received in revised form
2 May 2014
Accepted 5 May 2014
Available online 3 June 2014
Keywords:
Radiography
Dose
Risk
Thermoluminescent dosimetry
Monte Carlo method
* Corresponding author. Tel.: þ44 1612956414.
E-mail address: [email protected] (A. Tootel
http://dx.doi.org/10.1016/j.radi.2014.05.002
1078-8174/� 2014 The College of Radiographers. Pub
a b s t r a c t
Purpose: This paper gives an overview of the methods that are
used to calculate dose and risk from
exposure to ionizing radiation as a support to other papers in
this special issue.
Background: The optimization of radiation dose is a legal
requirement in medical exposures. This review
paper aims to provide the reader with knowledge of dose by
providing definitions and concepts of
absorbed, effective and equivalent dose. Criticisms of the use of
effective dose to infer the risk of an
exposure to an individual will be discussed and an alternative
approach considering the lifetime risks of

cancer incidence will be considered.
Prior to any dose or risk calculation, data concerning the dose
absorbed by the patient needs to be
collected. This paper will describe and discuss the main
concepts and methods that can be utilised by a
researcher in dose assessments. Concepts behind figures
generated by imaging equipment such as dose-
area-product, computed tomography dose index, dose length
product and their use in effective dose
calculations will be discussed. Processes, advantages and
disadvantages in the simulation of exposures
using the Monte Carlo method and direct measurement using
digital dosimeters or thermoluminescent
dosimeters will be considered.
Beyond this special issue, it is proposed that this paper could
serve as a teaching or CPD tool for
personnel working or studying medical imaging.
� 2014 The College of Radiographers. Published by Elsevier
Ltd. All rights reserved.
Introduction
Within this special issue of radiography, several articles use
Monte Carlo mathematical methods to estimate radiation dose to
humans. It is appreciated that readers may have a limited
knowl-
edge of these methods, or indeed measurement methods which
are
used to estimate dose. For readers with limited knowledge in
dose
estimation this article explains a range of approaches which
might
be used. This article also outlines concepts and defines terms
associated with dose and it discusses how data can be used to
provide an indication of risk to an individual.

Ionising radiation is made up of sub-atomic particles or, in the
case of X-rays and gamma rays, it comprises electromagnetic
waves
from the high energy part of the electromagnetic spectrum. At
energies associated with medical imaging, these particles and
waves have sufficient energy to ionise an atom and liberate an
electron. This process may lead to tissue damage which can
result
in cell mutation or apoptosis. The higher the dose of radiation,
the
l).
lished by Elsevier Ltd. All rights re
greater chance that tissue damage will occur.1 This probability
model of biological damage is referred to as the stochastic
effect. It
suggests that no dose of radiation is safe. It is this worst case
sce-
nario that radiation protection is based on, in that operators
should
aim to minimise the probability of tissue damage by using the
least
practicable amount of ionising radiation.2
In medical imaging a range of professionals are responsible for
ensuring doses are as low as reasonably practicable (ALARP).
The
operator performing the exposure is required to have an under-
standing of the steps that they can take to optimise dose thus
minimising the chance of stochastic affects.
Absorbed dose
Interactions of ionising radiation with matter can result in a
proportion of radiation energy being deposited. The amount of
energy deposited per unit mass is the absorbed dose

(represented
by the letter D) and is defined as joules per kilogram (J kg�1).
The SI
unit of absorbed is the Gray (Gy). Quantities of absorbed dose
are
usually quoted as milli-Gray (mGy, 1/1000 of a Gray) or a
micro-
Gray (mGy, 1/1,000,000 of a Gray).
served.
mailto:[email protected]
http://crossmark.crossref.org/dialog/?doi=10.1016/j.radi.2014.0
5.002&domain=pdf
www.sciencedirect.com/science/journal/10788174
http://www.elsevier.com/locate/radi
Table 1
Recommended radiation weighting factors from ICRP 103.1
Radiation type Radiation weighting factor
Photons (X-ray and gamma ray) 1
Electrons 1
Alpha particles 20
Protons 2
Table 2
Tissue weighing factors from ICRP 103.1
Organ Tissue weighting factor
Gonads 0.08

Bone marrow 0.12
Colon 0.12
Lung 0.12
Stomach 0.12
Breast 0.12
Bladder 0.04
Liver 0.04
Oesophagus 0.04
Thyroid 0.04
Skin 0.01
Bone (surface) 0.01
Salivary glands 0.01
Brain 0.01
Remaindera 0.12
Total 1.00
a Remainder tissues: adrenals, extrathoracic (ET) region, gall
bladder,
heart, kidneys, lymphatic nodes, muscle, oral mucosa, pancreas,
pros-
tate (_), small intestine, spleen, thymus, uterus/cervix ().
A. Tootell et al. / Radiography 20 (2014) 323e332324
Equivalent dose
The chance of tissue damage occurring does not just depend on
the absorbed dose but also the type and energy of the radiation.
Equivalent dose (represented by the symbol H) takes these
factors
into consideration and is obtained by applying a radiation
weight-
ing factor (W) to the absorbed dose. Radiation weighting factors
are
published by The International Commission On Radiation
Protec-

tion (ICRP)1; they reflect the biological damage potential of
different
radiation types (Table 1). It can be considered a less
fundamental
quantity than absorbed dose but it is useful for indicating the
health
risk of radiation exposure. Equivalent dose is still defined as
joules
per kilogram, but is assigned the SI unit Sievert (Sv). Figures
are
often quoted as milli-Sieverts (mSv) or micro-Sieverts (mSv).
The equation for equivalent dose is defined in Fig. 1.
Ionising radiation that forms part of the electromagnetic spec-
trum (e.g. X-ray and gamma ray) ionise atoms through the
photo-
electric absorption and the Compton effect. Both these
interactions
will eject an electron from an atom; this electron may ionise
many
more atoms. Since most of the affected atoms are ionised
indirectly
by the secondary electrons, photons are considered to be
indirectly
ionising.
Using the data in Table 1, it can be seen that an absorbed dose
of
1 mGy of X-ray photons results in an equivalent dose of 1 mSv,
i.e.
1 mGy � 1 (radiation weighting factor for X-ray photons),
where
1 mGy of alpha particles results in an equivalent dose of 20
mSv, i.e.
1 mGy � 20 (radiation weighting factor for alpha particles). In

other
words alpha particles have a higher risk to biological tissue.
Effective dose
Effective dose (represented by the symbol E) takes into account
the type and amount of exposed tissue. Different tissues within
the
body have difference sensitivities to radiation meaning a dose
applied to one area of the body can carry a higher risk than the
same dose applied to another. Effective dose takes the
equivalent
doses of a number of organs and through the application of a
tissue
weighting factor, the sum of these aims to provide a single
number
that is proportional to the detriment from a particular exposure.
It
allows comparisons of the risks associated with different
imaging
techniques or modalities.
The tissue weighting factors (Table 2) represent the sensitivity
of
their respective tissue, for example bone marrow is highly
sensitive
to radiation and so has a weighting factor of 0.12 where the
brain is
less sensitive and so has a weighting factor of 0.01. The sum of
the
tissue weighting factors is 1 and so the sum of the weighted
equivalent doses would provide a whole body effective dose.
Per-
forming this process for different techniques allows for a
compar-
ison of doses and an indication of the detriment of these.

Figure 1. Equation for calculating equivalent dose from
absorbed dose.1
Effective dose is still defined as joules per kilogram and also
has
the same SI unit as equivalent dose, Sievert (Sv, with figures
quoted
as milli-Sieverts (mSv) or micro-Sieverts (mSv)). The equation
for
calculating effective dose is shown in Fig. 2.
Controversies with effective dose
Effective dose is commonly used in medical imaging to compare
the risks from different modalities (for example, CT of the
cervical
spine versus conventional radiographic imaging of the same
anatomical area) or examinations that have differing dose distri-
butions (for example, comparison of effective dose from an
antero-
posterior hip to that from an antero-posterior shoulder). The
application of the tissue weighting factors to the equivalent
doses
of the organs provides the whole body effective dose from that
exposure. The application of effective dose is useful in these
situ-
ations as it provides referrers, practitioners and operators with
data
that allows them to make decisions during the referral,
justification
and optimisation of medical imaging procedures.2,3 When used
for
this purpose effective dose is a useful figure to use, however a
number of publications use this figure to calculate the risk of
the
exposure to an individual. As noted by a number of authors, and
the

ICRP themselves, the effective dose concept is not intended to
be
used this way as a number of factors are not taken into
consider-
ation.3e8
The tissue weighting factors are averaged over all ages and both
genders in the general population and so it cannot be applied to
an
individual patient.9 For example a measurement of organ doses
and
effective dose calculations from a chest radiograph could be the
same in a 15-year-old and a 35-year-old female. Using data pub-
lished by Wall et al.9 the lifetime risk of cancer incidence for
breast
tissue in 10e19 year olds of 3.34% per Gray and 30e39 year olds
of
1.44% per Gray it can be seen that the risk to female breast in
10e19
Figure 2. Equation for the calculation of effective dose.1
Table 3
Comparison of the tissue weighting factors form ICRP
publications 26, 60 and
103.1,12,13
Organs Tissue weighting factors
ICRP 26
(1977)12
ICRP 60
(1990)13

ICRP 103
(2007)1
Gonads 0.25 0.20 0.08
Red bone marrow 0.12 0.12 0.12
Colon e 0.12 0.12
Lung 0.12 0.12 0.12
Stomach e 0.12 0.12
Breasts 0.15 0.05 0.12
Bladder e 0.05 0.04
Liver e 0.05 0.04
Thyroid 0.03 0.05 0.04
Skin e 0.01 0.01
Bone surface 0.03 0.01 0.01
Salivary glands e e 0.01
Brain e e 0.01
Remainder 0.03 0.05 0.12
Total 1.00 1.00 1.00
Table 4
Life time risks of cancer incidence for males and females by
organ and age for a Euro-
American population (% per Gy).
Organ (male) Age at exposure (y)
0e9 10e19 20e29 30e39 40e49 50e59 60e69
Lung 0.65 0.69 0.73 0.78 0.80 0.76 0.61
Stomach 0.93 0.73 0.57 0.43 0.31 0.20 0.12
Colon 1.49 1.22 0.98 0.79 0.60 0.43 0.25
Red bone marrow 1.06 1.05 0.77 0.76 0.78 0.65 0.49
Bladder 0.89 0.76 0.65 0.55 0.46 0.35 0.23
Liver 0.56 0.44 0.34 0.26 0.18 0.12 0.07
Thyroid 0.18 0.10 0.05 0.03 0.01 0.01 0.00

Oesophagus 0.12 0.11 0.11 0.11 0.12 0.14 0.15
Organ (female) Age at exposure (y)
0e9 10e19 20e29 30e39 40e49 50e59 60e69
Breast 4.92 3.34 2.21 1.44 0.84 0.45 0.21
Lung 1.36 1.46 1.58 1.70 1.78 1.72 1.39
Stomach 1.45 1.14 0.88 0.67 0.48 0.33 0.20
Colon 0.73 0.59 0.48 0.38 0.29 0.21 0.14
Red bone marrow 0.48 0.48 0.50 0.45 0.77 0.49 0.29
Bladder 0.70 0.61 0.52 0.45 0.39 0.32 0.24
Liver 0.24 0.19 0.15 0.11 0.08 0.06 0.03
Thyroid 0.92 0.52 0.26 0.13 0.06 0.02 0.01
Oesophagus 0.10 0.09 0.10 0.12 0.15 0.21 0.28
A. Tootell et al. / Radiography 20 (2014) 323e332 325
year old is higher. This difference in sensitivity due to age and
gender is not captured within conventional effective dose
calculations.
The tissue weighting factors published by the ICRP are derived
using data that is assessed and analysed by The United Nations
Scientific Committee on the Effects of Atomic Radiation
(UNSCEAR)
on cancer risks from follow-up studies of the Japanese atomic
bomb
survivors.10 As a result of these on going long-term studies the
tissue weighting factors have undergone a number of revisions
as
new data on cancer incidence has been collected.11 The effect
of
these revisions can be seen in Table 3.
From Table 3, it can be seen that weightings assigned to the
gonads have undergone significant changes over the three publi-

cations, from 0.25 in ICRP 26 to 0.08 in ICRP 103 that is a
reflection
of the understanding of heritable risk and the change in breast
tissue changed from 0.05 in 1990 to 0.12 in 2007 due to a
decision
by the ICRP committee to put more emphasis on cancer
incidence
rather than mortality.4e6 Brenner, in a number of publications,
suggests that these tissue weighting factors represent a
subjective
balance between the different stochastic endpoints of cancer
inci-
dence, cancer mortality, life shortening and hereditary risk. This
subjectivity is an example of the “flaws in the science” behind
the
derivation of these factors.4e6 However, Dietze11 argues that
this
revision was in response to the publication of more reliable
cancer
incidence data published by UNSCEAR10 rather than a change
in the
committee’s emphasis. Whatever the reason, it is clear that
these
revisions do have an impact on effective dose calculations
making
comparisons to older data difficult.
Lifetime risk of cancer induction
It is with these criticisms in mind that Brenner proposes an
alternative to effective dose that can be applied to individual
Figure 3. Equation for calculating effective risk.2e4
patients - this is referred to by Brenner as “effective risk”.
Effective
risk considers the life time risk of cancer induction from an
absorbed dose of radiation and the equation for this is shown in

Fig. 3.4
This equation is very similar to that used to calculate effective
dose. It is proposed that the tissue weighting factors are
replaced
with organ-specific radiation-induced cancer risk, such as those
published by The Nuclear and Radiation Studies Board14 or
more
recently by Wall et al.9 (a selection of data is shown in Table
4).
The lifetime risk figures (Table 4) are calculated from stronger
data as they are based directly on epidemiological studies and
not
decided by committee.10,14 The organ-specific radiation-
induced
risk data reflects current knowledge in the biological effects of
ra-
diation. The results would be easier to interpret (for example x
per
1,000,000) for medical imaging professionals and non-medical
imaging professionals too. The lifetime risk can be used to pro-
vide risk to different genders and age groups.
Conveying risk to patients is arguably one of the more chal-
lenging aspects the radiography profession has to contend with.
To
this end, Wall suggests using a category based approach to
convey
the risk from the radiological examination (Table 5).9
Prior to any analysis of risk, dose data has to be collected. Dose
data can be measured or estimated. In medical imaging either
method can be used to determine dose data in most situations,
although there could be occasions where only one method is
suitable.

Modelling dose
Mathematical modelling of dose using commercially available
software is relatively quick and easy. Software is available that
Table 5
Four broad risk bands for the typical total lifetime cancer risk
for patients.9
Category Total lifetime cancer risk
Negligible risk Less than 1 in a million
Minimal risk 1 in a million to 1 in 100,000
Very low risk 1 in 100,000 to 1 in 10,000
Low risk 1 in 10,000 to 1 in 1000
allows for organ and effective dose values for conventional
radio-
graphic techniques and CT imaging to be estimated. The
software
employs Monte Carlo modelling which is a mathematical
technique
that simulates as closely as possible the real interactions
suffered by
photons.
The process involves the computer simulation of an anthropo-
morphic phantom being exposed to a large number of photons of
varying energies emitted from a point source. The path of each
photon is followed through a sequence of interaction points and
subsequent energy losses and outgoing directions (through
coherent scattering or Compton scattering). This chain of in-
teractions forms a so-called photon history. At each interaction

point the energy deposited to the organ is calculated and used in
the dose calculation. A large number of independent random
photon histories are generated and estimates of the mean values
of
the energy depositions in the various organs are used for calcu-
lating the dose in these organs Eventually, the photon loses
suffi-
cient energy to allow photoelectric absorption to occur.15,16
PCXMC (STUUK, Helsinki, Finland16,17) is one such
programme
that allows for organ and effective dose to be estimated in many
conventional radiographic techniques. Fig. 4 illustrates the
posi-
tioning of a PA chest with landscape orientation of the image
re-
ceptor. Other parameters can be manipulated in the software
including X-ray anode angle, tube filtration material and
thickness
to obtain final dose estimates.
Figure 4. Example of data entry page of PCXMC for the calculat
Dose modelling software is also available for CT dose estima-
tions. For example, ImPact’s CT Dosimetry Tool (ImPact, Lon-
don18,19) software simulation allows for quick and easy
calculation
of organ and effective dose through the use of Monte Carlo data
for
normalised organ doses. However, as can be seen in Fig. 5,
results
are dependent on selecting the imaging parameters and CT
model
as calculations take into account specific features of each CT
unit
(e.g. radiation quality and field geometry).20 Selection of the
correct
scanner may not always be possible as new technology and

systems
are constantly being introduced. These systems are currently not
included in dose simulation software meaning that dose
simulation
has to rely on “best fitting” the attributes of these scanners to
those
of a similar design. As noted by Groves et al.,21 this introduces
the
potential for significant error in the estimated doses. Automatic
mA
manipulation by the scanners can also lead to error as the
software
only allows a single value to be used.
Underestimation of CT doses using computer simulation is
frequently reported with magnitudes between 18 and
40%.5,12,13
Reasons for these underestimations have been explained by the
differences in the physical dosimetry phantoms and the virtual
phantoms used by dose modelling software. Close examination
of
this highlights the simplified geometric shapes of the organs.
Subsequently, as can be seen in Fig. 6, in CT examinations of
the
chest the CT virtual phantom suggests that the liver is not
exposed
to primary X-ray beam and thus the calculated liver dose would
be
ion of organ and effective dose from a PA chest radiograph.
Figure 5. Example of the dose report generated by ImPact CT
Dosimetry software.

low. In reality a significant volume of this organ is included in
the
scan and so will contribute to the effective dose calculations.
However, accuracy of CT dose modelling can be improved by
careful selection of the scan range to match the fractions of
organs
irradiated and to include overbeaming and overranging that is a
feature of helical scanning. The use of an average mAs for the
scan
parameters will further improve the accuracy of the dose
calculations.22
Figure 6. Comparison of the virtual phantom in ImPact CT
dosimetry software and ATOM dosimetry phantom.
Figure 7. Typical arrangement of the phantom and pencil beam
ionisation chamber
used to collect CTDI.
Measuring dose
There are many tools which may be used to measure the dose
absorbed during a radiographic procedure that will allow dose
to be
calculated. The one that most operators will be familiar with is
the
dose area product (DAP) meter. DAP combines two quantities e
as
its name suggests absorbed dose in air and the field size giving
the
unit Gray centimetre squared Gycm2 (or cGycm2 or mGycm2)

(NB
not Gray per square centimetre). DAP meters are mounted onto
the
X-ray tube in front of the collimators making readings easy to
ac-
quire, however it is important to note that DAP is not patient
dose
per se.23 It is independent of the distance between the source
and
the patient meaning that if this figure is to be used to estimate
patient dose the source to patient distance, the field size and the
location of the area exposed are required.24
CT acquisitions have similar values that are often used as a
reference for patient dose; the computed tomography dose index
(CTDI) and dose length product (DLP). The CTDI measurement
was
based on an axial CT scanner and was defined as the dose from
the
primary beam plus scatter from surrounding slices from a single
slice in an acrylic phantom (Fig. 7). Phantoms come in two di-
ameters, 16 cm and 32 cm, to represent the head and body
respectively.25
Developments in technology and the advent of multislice CT
equipment lead to variations in CTDI. CTDI100 reflects the
dose
contribution from a 100 mm range (50 mm either side of the
reference slice). The weighted CTDI (CTDIw) reflects the
weighted
sum of two-thirds the peripheral dose and one-third the central
dose in a 100 mm range. The most commonly quoted CTDI
value in
modern CT technology is the volume CTDI (CTDIvol). This
value is
obtained by dividing CTDIw by the beam pitch factor.

22,26 As before
CTDI in any form is not patient dose but a quantification of the
radiation output of the CT system so does not take into account
differing patient sizes and area of the body that is being
imaged.27
A derivative of CTDI is dose length product (DLP). This figure
takes into account the length of the scan and is calculated by
multiplying the CTDIvol by the length of the scan. In a similar
way to
CTDI and CTDIvol, DLP is not patient dose as it does not take
into
account what part of the body is being exposed, the size of the
patient, or the patient’s age.
Conversion of DLP to patient dose is possible using a
conversion
coefficient (k) shown in (Table 6). This conversion factor is
defined
as the effective dose per dose-length product and has the unit
mSv/
mGy cm. Multiplying the DLP by the relevant conversion factor
gives a value for effective dose.
Criticisms of k state that the factors are based on old
technology
and old data; they are based on several scanners that were in use
circa 1990 and the tissue weighting factors used in their
calculation
are from ICRP 60.13,26,28 There are also a number of
assumptions
made that would increase the error in the calculated effective
dose.
For example, the patient is assumed to be standard, and as noted
by

McCollough et al.,27 this standard patient is a little thin by
today’s
Table 6
Normalised values of effective dose per dose-length product
(DLP) over various body
regions.
Region of the body Normalised effective dose
(E/DLP) (mSv/mGycm
Head 0.0023
Neck 0.0054
Chest 0.017
Abdomen 0.015
Pelvis 0.019
Figure 9. Lower thoracic slices of a paediatric phantom showing
different density
resin for the lung, soft tissue and bone with locations for
dosimeters. These dosimeters
can be electronic (shown here by the two wires inserted into the
phantom) or
analogue such as thermoluminescent dosimeters.29
standards (nominal body mass of 70 kg). Variation in the way
CT
scanners report CTDIvol for paediatric patients can make
compari-
son difficult. Some use the 16 cm phantom while others use the
32 cm phantom. For example Siemens, Philips dose reports are
based on a 32 cm phantom, Toshiba reports are based on 16 cm
phantom and GE reports use the 16 cm or 32 cm depending on

the
scan field of view. CTDIvol can differ by a factor of
approximately 2.5
between the two diameter phantoms.27
True measurement of dose using digital or analogue dosimeters
such as metal oxide semi-conductor field effect transistor
(MOS-
FET) or thermoluminescent dosimeters (TLDs) (described later)
can
be done in a number of ways. In the experimental setting it is
possible to measure organ dose by placing dosimeters in a
specially
designed anthropomorphic phantom. These phantoms are avail-
able in a range of patient types; male and female and paediatric,
adolescent and adult (Fig. 8). They are made up of contiguous
slices
with different tissues represented by different densities of
epoxy
resin. The resin has attenuation properties that are equivalent to
real tissue. Within the slices are locations for placing
dosimeters
that will provide data of organ dose (Fig. 9). Using these
phantoms
allows the researcher to carry out experimentation on different
techniques, exposure factors or positioning to optimise dose
without the involvement of real patients.
Figure 8. The CIRS ATOM dosimetry phantom fami
It is obviously impossible to directly measure organ dose in the
clinical setting, so the entrance surface dose (ESD) can be used.
ESD
is defined as the absorbed dose in the skin at a given location
on the
patient and also includes backscattered radiation from the
patient.
As a measurement it can be combined with DAP to allow

calcula-
tions of patient dose to be made.
Dosimeters
ESD and organ dose in the anthropomorphic phantom can be
measured using a digital dosimeter or using thermoluminescent
ly models 701e706 (CIRS, Norfolk, Virginia).29
dosimeters (TLD). Most medical imaging personnel will be
familiar
with TLDs in the context of radiation protection as they are
frequently used in personal dosimeter badges. TLDs are
available in
a variety of forms, from powder to square or circular chips,
rods,
cubes and in a range of materials.
Thermoluminescence (illustrated in Fig. 10) uses the atomic
model of two energy bands; the valence band and conduction
band.
Within the valance band electrons are bound to individual atoms
as
opposed to the conduction band where electrons can move
freely
within the atomic lattice. Separating these two bands is an area
that
is referred to as the forbidden gap in which no electron state can
exist. The impurities mentioned above create electron traps
within
this gap. Exposure to ionising radiation excites electrons
allowing
them to move up to the conduction band leaving holes within

the
valence band. Electrons can travel amongst the crystal lattice
until
either the electron can cross back towards the valence band and
recombine with a hole or, if near a defect, it can fall into the
energy
trap. The electron is now prevented from filling a hole within
the
valance band until it can gain enough energy to once again
reach
the conduction band before moving back to the valance band.
This
stimulation is in this context accomplished by introducing
heat.30
The movement of the electron back to the valance band requires
the electron to lose energy. This energy is released in the form
of
visible light and this light is detected by a photomultiplier tube.
The
charge (measured in Coulombs [C]) generated from this
component
is measured.
Conduc on
Valance Band
TLD
TLD readout
Thermal s mula on
Conduc on
Valance BandTLD

Exposure to radia on
Conduc on
Valance Band
TLD
Storage
Emission of light
Figure 10. Illustration of the process involved in TLD
dosimetry.
The choice of material depends on the nature of radiation; in
diagnostic and therapeutic energies the chemical composition of
the dosimeters is either lithium fluoride with magnesium and ti-
tanium impurities added or lithium fluoride with magnesium,
copper and phosphorus impurities added.31 The difference in
the
materials affects their sensitivity and the measurement range the
TLD is capable of. For example TLDs made from Calcium
Fluoride
Dypromsium are suitable for environmental monitoring and as
capable of detecting doses of between 0.1 pGy and 1 Gy.32
Lithium
Fluoride with magnesium and titanium are suitable for medical
physics dosimetry applications and operate at doses between
10 pGy and 10 Gy.33
Conversion from charge to dose involves a calibration process.
The TLD or batches of TLDs plus scattering material and a
digital
dosimeter are exposed to a range of exposures at energy (kV)
consistent with the experiment that will be performed. The

charge
generated from the reading process and the doses recorded by
digital dosimeter are used to establish the calibration factor
through linear regression. An example of calibration data using
is
shown in Fig. 11.
General radiographic equipment can be used in this process
although some TLD readers will perform calibration using
sealed
sources of gamma emitting isotopes such as Strontium-90 or
Yttrium-90. Such systems will calibrate each TLD individually
rather than in batches increasing the accuracy of the final
readings.
The response of the TLDs is energy dependent therefore
calibration
should be performed at the energy that will be used in the
research
or measurements. If this cannot be done then energy conversion
factors can be used but this can introduce error.34
One of the disadvantages of the TLD is the time needed to pre-
pare and setup and process them. A typical whole body adult
phantom measurement for calculation of effective dose involves
the use of 268 individual TLDs. Reading this number using a
manual
Figure 11. Example of calibration data for TLDs at 80 kV.
Plotting data from table (a)
results in graph (b) and shows the linear relationship between
the charge generated
from reading the TLD to dose. The gradient of this line is the
calibration factor.
Figure 12. A MOSFET dosimeter with an array of five

dosimeters connected to the
module.36
TLD reader equates to approximately 6 h of work.35 Research
has
been undertaken to follow the dental radiography dosimetry pro-
cess to reduce the number of TLD required for effective dose
measurement however, if comparison of organ dose and risk is
to
be carried out it has been shown that a measurement organ dose
is
required for all critical organs.35
An alternative to TLDs is the digital dosimeter. An example of
this is the metal oxide semi-conductor field effect transistor
(MOSFET) (Best Medical Canada, Ontario, Canada) (Fig. 12).
Exposure of the digital dosimeter results in a voltage shift
between the components of the dosimeter. This difference is
measured and is proportional to the dose absorbed by the de-
tector. However, it is unlikely that the total number of digital
dosimeters would be available to allow measurement of all
critical
organs in one exposure due to expense of the dosimeters.
Therefore a number of repeated measurements with the dosim-
eter relocated between each would be required. As with TLDs,
MOSFET dosimeters require calibration and their response is
en-
ergy dependent meaning separate calibrations are required if a
significant difference in beam energies is to be used in any
research37
Measurements using TLD or digital dosimeters have their ad-
vantages and disadvantages relating to preparation time,
acquisi-

tion time processing following exposure, and cost. These have
to be
considered when planning research that involves the direct mea-
surement of dose form exposure to ionising radiation.21,35,38
Summary
This article has given insight into terms and concepts associated
with dose measurement and modelling, as well as risk
estimation.
Some limitations and values of dose estimation and
measurement
methods have been considered. As support for this special issue
the
reader should have gained enough background and insight into
Monte Carlo mathematical dose modelling to be able to
appreciate
some of the empirical articles. Beyond the special issue we
antici-
pate that the article could serve as a teaching or CPD aid for
personnel working in medical imaging.
Conflict of interest statement
None.
References
1. ICRP. The 2007 recommendations of the ICRP, in ICRP
publication 103. Ann
ICRP; 2007:2e4.
2. The ionising radiation (medical exposure) regulations, United
Kingdom; 2000.
3. Oritz P. The use of effective dose in medicine. In: The
second international

symposium on the system of radiological protection. United
Arab Emirates: ICRP;
2013.
4. Brenner DJ. Effective dose: a flawed concept that could and
should be replaced.
Br J Radiology 2008;81(967):521e3.
5. Brenner DJ. Effective dose e a flawed concept that could and
should be replaced.
Washington, DC: International Commission on Radiological
Protection; 2011.
6. Brenner DJ. We can do better than effective dose for
estimating or comparing
low-dose radiation risks. Ann ICRP 2012;41(3, 4):124e8.
7. Harrison J. The use of effective dose. In: The second
international symposium on
the system of radiological protection. United Arab Emirates:
ICRP; 2013.
8. Martin CJ. Effective dose: how should it be applied to
medical exposures? Br J
Radiology 2007;80(956):639e47.
9. Wall BF, Haylock R, Jansen JTM, Hillier MC, Hart D,
Shrimpton PC. Radiation
risks from medical X-ray examinations as a function of the age
and sex of the
patient. Rep HPA-CRCE-028. Chilton: Health Protection
Agency; 2011.
10. United Nations. Scientific committee on the effects of
atomic radiation, effects of
ionizing radiation: United Nations Scientific Committee on the

effects of atomic
radiation e UNSCEAR 2006 report, volume II e report to the
General Assembly,
with scientific Annexes C, D, and E. UN; 2009.
11. Dietze G, Harrison JD, Menzel HG. Effective dose: a flawed
concept that could
and should be replaced. Comments on a paper by D J Brenner
(Br J Radiol
2008;81:521e3). Br J Radiology 2009;82(976):348e51.
12. ICRP. Recommendations of the ICRP. ICRP Publication 26
Ann ICRP; 1977:3.
13. ICRP. 1990 Recommendations of the ICRP. ICRP
Publication 60 Ann ICRP; 1991:
1e3.
14. Committee to Assess Health Risks from Exposure to Low
Levels of Ionizing
Radiation, Nuclear and Radiation Studies Board Division on
Earth and Life
Studies, and National Research Council of the National
Academies, Health Risks
From Exposure to Low Levels of Ionizing Radiation: BEIR VII
Phase 2,Wash-
ington, DC; 2006.
15. Baker C. Physicist toolbox e basics of Monte Carlo
simulation of radiation trans-
port. Liverpool: United Kingdom Radiological Congress; 2013.
16. Tapiovaara M, Siiskonen T. In: PCXMC: a Monte Carlo
program for calculating pa-
tient doses in medical X-ray examinations. 2nd ed. Helsinki,
Finland: STUK; 2008.

17. Tapiovaara M, Lakkisto M, Servomaa A. PCXMC 2.0
rotation dose calculations.
Helsinki, Finland: STUK; 2012.
18. ImPACT. Imaging performance assessment of CT scanners.
Available from:
http://www.impactscan.org [cited 22.04.2012].
19. ImPACT. CT patient dosimetry calculator. London: Medical
Devices Agency;
2003.
20. Hashemi-Malayeri BJ, Williams JR. A practical approach
for the assessment of
patient doses from CT examinations; 2003. Available from:
www.dundee.ac.
uk/medphys/documents/hashemi.pdf [cited 10.01.2012].
21. Groves AM, Owen KE, Courtney HM, Yates SJ, Goldstone
KE, Blake GM, et al. 16-
Detector multislice CT: dosimetry estimation by TLD
measurement compared
with Monte Carlo simulation. Br J Radiol 2004;77(920):662e5.
22. Castellano E. CT dose calculations for individual patients e
what you should
know. In 12th CT users group, London; 2010.
23. Conference of radiation control program directors
committee on quality
assurance in diagnostic X-ray, dose-area product (DAP), in Q.A.
collectible,
Frankfort; 2001 [reviewed and republished 2008].
24. Moores BM. Radiation dose measurement and optimization.

Br J Radiology
2005;78(933):866e8.
25. Coursey C, Frush D. CT and radiation: what radiologists
should know. Appl
Radiol 2008;37(3):22e9.
26. McNitt-Gray M. Assessing radiation dose: how to do it
right, in 2011 AAPM CT
dose summit, Denver; 2011.
27. McCollough CH, Leng S, Yu L, Cody DD, Boone JM,
McNitt-Gray MF. CT dose
Index and patient dose: they are not the same thing. Radiology
2011;259(2):
311e6.
28. European Commission, EUR 16262 EN. European guidelines
for quality criteria
for computed tomography. Luxembourg: European Commission;
1999.
29. CIRS tissue simulation and phantom technology. Dosimetry
verification
phantoms model 701e706 data sheet; 2012. Available from:
http://www.
cirsinc.com/file/Products/701_706/701_706_DS.pdf [cited
19.01.2012].
30. Stabin MG. Radiation protection and dosimetry: an
introduction to health physics.
Springer; 2007.
31. Bilski P. Lithium fluoride: from LiF:Mg,Ti to LiF:Mg,Cu,P.
Radiat Prot Dosim
2002;100(1e4):199e205.

32. TLD-200� thermoluminescent dosimmeter.
ThermoScientific; 2014. Available from:
http://www.thermoscientific.com/en/product/tld-200-
thermoluminescent-
dosimetry-material.html [cited 29.04.2014].
33. TLD-100� thermoluminescent dosimetry material.
ThermoScientific;2014. Available
from: http://www.thermoscientific.com/en/product/tld-100-
thermoluminescent-
dosimetry-material.html [cited 29.04.2014].
34. Podgor�sak EB, International Atomic Energy Agency.
Radiation oncology physics:
a handbook for teachers and students. International Atomic
Energy Agency;
2005.
http://refhub.elsevier.com/S1078-8174(14)00056-X/sref1

http://www.impactscan.org
http://www.dundee.ac.uk/medphys/documents/hashemi.pdf
http://www.dundee.ac.uk/medphys/documents/hashemi.pdf

ThompsonEtal2013.pdfAccurate localization of incidental fi.docx

ThompsonEtal2013.pdfAccurate localization of incidental fi.docx

Recommended

Recommended

More Related Content

Similar to ThompsonEtal2013.pdfAccurate localization of incidental fi.docx

Similar to ThompsonEtal2013.pdfAccurate localization of incidental fi.docx (20)

More from herthalearmont

More from herthalearmont (20)

Recently uploaded

Recently uploaded (20)

ThompsonEtal2013.pdfAccurate localization of incidental fi.docx