Medical Imaging and Radiotherapy Journal (MITRJ) 37 (1) 25
Original article
IMPACT OF ACQUISITION PARAMETERS ON THE QUANTITATIVE 
ASSESSMENT OF PET IMAGING – ANALYSIS OF THE NEMA 
PHANTOM
VPLIV SLIKOVNIH PARAMETROV NA KVANTITATIVNO OCENO PET SLIKE – 
ANALIZA FANTOMA NEMA
Sebastijan Rep
University Medical Centre Ljubljana, Clinic of Nuclear medicine, Zaloška 7, 1000 Ljubljana, Slovenia
*Corresponding author: sebastijan.rep@guset.arnes.si
Received: 7. 4. 2020
Accepted: 25. 7. 2020
https://doi.org/10.47724/MIRTJ.2020.i01.a005
IZVLEČEK
Namen: Analizirati najpogostejše dejavnike, ki vplivajo na 
vrednosti SUV. 
Materiali in metode: V raziskavi sem uporabil fantom NEMA, 
napolnjen z mešanico vode in 18F-FDG v razmerju 1:4 (ozadje/
sfere) in analiziral najpogostejše dejavnike, ki vplivajo na 
vrednost SUV. Najpogostejši dejavniki vključujejo vpliv telesne 
teže pacienta, vpliv časa med aplikacijo in slikanjem s PET/
CT in vpliv različno pripravljene in aplicirane koncentracije 
aktivnosti radiofarmaka (RF).
Rezultati: Različne vrednosti telesne teže pacienta, čas med 
aplikacijo in slikanje s PET/CT in različno pripravljene in 
aplicirane koncentracije aktivnosti RF statistično pomembno 
vplivajo na kvantitativno oceno SUVmax (p < 0,001 in 
SUVmean (p < 0,001).
Zaključek: Rezultati so pokazali, da lahko vsi dejavniki 
pomembno vplivajo na kvantitativno oceno SUVmax in 
SUVmean.
Ključne besede: PET/CT, kvantitativna ocena, SUVmax, 
SUVmean, telesna teža
ABSTRACT
Aim: The aim of the research was to analyse the most common 
factors that infl uence SUV values. 
Material and methods: In the study, I used a NEMA body 
phantom fi lled with a mixture of water and 18F-FDG in a ratio 
1:4 (background/spheres), and analysed the most common 
factors that infl uence SUV values. The most common factors 
include the impact of the patient's body weight, the impact 
of time between application and PET/CT imaging, and the 
impact of diff erently prepared and administered RP activities.
Results: Diff erent values of patient body weight, time between 
application and PET/CT imaging, and diff erently prepared and 
administered RF activities have a statistically signifi cant eff ect 
on the quantitative assessment of SUVmax (p < 0.001) and 
SUVmean (p < 0.001).
Conclusion: The results showed that all factors can 
signifi cantly infl uence the quantitative assessment of SUVmax 
and SUVmean.
Keywords: PET/CT, quantitative assessment, SUVmax, 
SUVmean, body weight
 
26 Medical Imaging and Radiotherapy Journal (MITRJ) 37 (1)
INTRODUCTION
Positron emission tomography (PET) in combination with 
computed tomography (CT) is a hybrid imaging diagnostic 
method that is frequently used for diagnosis, prognosis and 
monitoring response to oncological therapy. The hybrid 
system facilitates a parallel anatomical image using computed 
tomography (CT) and functional image using PET.  Visual 
assessment is the main tool for image interpretation in clinical 
practice. Although visual assessment may be suffi  cient to 
evaluate tumour response, a precise assessment of tumour 
response to therapy requires a certain form of quantifi cation 
(1, 2). PET is a diagnostic imaging method that facilitates 
the quantitative assessment of the pathological process. 
The quantitative assessment enables objective and precise 
evaluation to predict and monitor therapeutic response 
so that it does not depend solely on the visual imaging 
assessment. Quantifi cation in PET examinations represents an 
accumulated amount of radiopharmaceuticals (RP) inside the 
tumour and facilitates a precise division into groups of patients 
who experience therapeutic response and those who do not 
(3, 4). A quantitative analysis utilising 18F- fl uorodeoxyglucose 
(18F-FDG) in the assessment of early therapeutic response 
increased the role of PET in drug development in oncology (5). 
Standardised uptake value (SUV) is a simplifi ed quantitative 
assessment and the most frequently used method to assess 
accumulation of 18F-FDG PET in examinations (6, 7).
SUV is a number that stands for the accumulation activity of 
RF inside the tumour or the entire body that was measured 
after intravenous RP application in a predetermined period. 
SUV is normalised to the applied RF dose and factor that takes 
into account the distribution of RF in the whole body (8, 9). 
The most common factors for normalisation of RF distribution 
in the whole body are body weight (BW) and body surface 
area (BSA). Patients who undergo an 18F-FDG exam must do 
so on an empty stomach. Those patients have a decreased RP 
accumulation in the fat, which can impact body weight, and 
therefore, a method considering lean body mass (LBM) is used. 
LBM is defi ned as the diff erence between total body mass 
and body fat, and takes into account the mass of all organs, 
excluding body fat. The use of LBM for the normalisation of 
SUV is more appropriate for heavier patients than BW or BSA 
(10, 11).
Physiological and technical factors or factors that are the 
result of human error impact the results of the highest 
concentration activity (SUVmax) and average concentration 
activity (SUVmean). A common technical error that impacts 
SUV is a discrepancy in time at PET/CT and dose meter. To 
avoid an incorrect SUV, time on a dose meter and PET/CT 
should be checked daily. Data collection period is an image 
parameter that impacts the signal-to-noise ratio (SNR) and 
consequently SUV. Along with the processing parameters, 
SUV is also impacted by the selection of the matrix element 
(due to the eff ect of partial volume), reconstruction algorithm 
and normalisation factor. SUV is also dependent on the region 
of interest (ROI), the selection of which is impacted by the 
individual’s choice.  Most common factors that impact SUV are 
shown in Table 1 (8, 12-14). 
Rep S. / Impact of acquisition parameters on the quantitative assessment of pet imaging – analysis of the nema phantom
Table 1: Most common factors that impact the quantitative as-
sessment of a PET image
CAUSE FACTOR EXPLANATION
Technical 
error
Incorrect time 
synchronisation 
between PET/CT and 
dose meter
Incorrect SUV due to 
erroneous correction of 
RP decay
Paravenous 
application of 
18F-FDG
Amount of applied RF 
is decreased, resulting 
into incorrect SUV
Physical 
factors
Imaging parameters Low SNR value causes a 
biased SUV 
Reconstruction 
parameters
Partial volume eff ect on 
SUV
Selection of ROI SUV result is highly 
dependent on the 
selection of ROI size
Normalisation factor 
for SUV
SUV depends on weight, 
body surface and other 
normalisation factors for 
the calculation of SUV
AIM
The aim of the study was to demonstrate and analyse the 
most common factors that impact SUVmax and SUVmean 
values. The following factors were included in the analysis: 
- impact of patient’s body weight, 
- impact of time between the application and data collection 
with PET/CT, and 
- impact of diff erently prepared and applied RF concentration 
activity.
METHODS
A NEMA body phantom was used to analyse factors that 
impact the quantitative assessment of SUV (SUVmax/mean). I 
conducted the phantom imaging on a SIEMENS hybrid system, 
Biograph mCT® 128 PET/CT, which combines a 128-slice CT 
and LSO PET detector system with three rings. The phantom 
volume was 9.7 litres and consisted of six hollow spheres 
with diameters of 37, 28, 22, 17, 13 and 10 mm. The phantom 
was fi lled with a mixture of water and 18F-FDG in a 1:4 ratio 
(background/sphere). When performing the PET/CT imaging, 
I collected the CT data for attenuation correction fi rst, then 
the PET data with the application of a single bed position. For 
the reconstruction of data, I used the iterative reconstruction 
algorithm (TrueX + TOF) that encompasses the point spread 
function (PSF) and time-of-fl ight information (TOF). SUV 
is most frequently normalised to body weight (BW) and is 
calculated using the following formula (Formula 1):
ACvoi (MBq ⁄ ml)
FDGdose (MBq) ⁄ BW (kg)
SUVBW=              Formula 1
In Formula 1, ACvoi represents the average or highest activity 
concentration expressed in MBq/ml, in a defi ned region of 
Medical Imaging and Radiotherapy Journal (MITRJ) 37 (1) 27
Rep S. / Impact of acquisition parameters on the quantitative assessment of pet imaging – analysis of the nema phantom
interest (ROI). FDGdose represents the dose of FDG expressed 
in megabequerels (MBq), while body weight (BW) is expressed 
in kilograms. 
SUV corrected to LBM is calculated using Formula 2:
ACvoi represents the average or highest activity concentration 
expressed in MBq/ml, in a defi ned region of interest (ROI). 
FDGdose represents the dose of FDG expressed in MBq. LBM 
value depends on the gender, body weight and height of a 
patient, and is calculated diff erently for men and women. LBM 
(women) = (1.07 x body weight) (kg) - 148 [body weight (kg) / 
body height (cm)]2 and LBM (men) = (1.1 x body weight) (kg) - 
128 [body weight (kg)/body height (cm)]2 (15).
SUV corrected to BSA is calculated using Formula 3:
 
 
ACvoi represents the average or highest activity concentration 
expressed in MBq/ml, in a defi ned region of interest (ROI). 
FDGdose represents the dose of FDG expressed in MBq. The 
height and weight of patients must be entered in the imaging 
or processing protocol to calculate BSA. The entered data are 
applied to calculate BSA, using the following formula: 
BSA (m2) = 0.007184 x body weight (kg)0.425 x body height 
(cm)0.725 (16).
I analysed the impact of incorrectly entered body weight, RP 
application time and prepared RP activity on SUVmax and 
SUVmean on a NEMA phantom. I systematically entered data 
in the imaging protocol and thus simulated an error. I altered 
the weight of the phantom (10 kg) by 10%, 20%, 30% and 40%, 
and analysed SUVmax and SUVmean. In terms of time impact 
on SUVmax and SUVmean, I simulated an error by entering 
times of 1, 3, 7, 15, 30, 45 and 60 minutes. By altering activity 
by 5% from the reference, I calculated the impact of lower and 
higher activity of prepared RF on the quantitative assessment 
of SUVmax and SUVmean.
On the images, I marked regions of interest of approximately 
six spheres of diff erent sizes in the NEMA body phantom. An 
analysis of obtained quantitative assessments of SUVmax/
mean was conducted using SPSS 25 software. The Shapiro-
Wilk test was applied to assess the distribution of variables. 
I used the analysis of variance (ANOVA) test (repeated 
measures) for dependent variables in the normal distribution 
and the Friedman test when variables were not distributed 
normally. I used a p value of < 0.05 for the threshold of 
statistical signifi cance.
RESULTS
An analysis of the normalisation of SUVmax and SUVmean to 
BW, LBM in BSA showed a statistically signifi cant diff erence at 
SUVmax (p = 0.002) and at SUVmean (p < 0.001). The obtained 
SUVmax and SUVmean values are shown in Tables 2 and 3.
Table 2: SUVmax values at diff erent sphere volumes normalised to 
BW, LBM and BSA
SUVmax value
Sphere volume BW LBM BSA
0.5 2.67 13.97 0.41
1.13 4.49 23.45 0.70
2.5 6.07 31.67 0.94
5.02 6.05 31.60 0.94
11.01 5.75 30.05 0.89
23.41 5.59 29.16 0.87
Table 3: SUVmean values at diff erent sphere volumes normalised to 
BW, LBM and BSA
SUVmean value
Sphere volume BW LBM BSA
0.5 2.4 12.53 0.37
1.13 3.15 16.45 0.47
2.5 3.57 23.56 0.55
5.02 4.08 21.08 0.63
11.01 4.26 22.24 0.66
23.41 4.57 23.86 0.71
ACvoi (MBq ⁄ ml)
FDGdose (MBQ) ⁄ LBM (kg)
ACvoi(MBq ⁄ ml)
FDGdose (MBq) ⁄ BSA (m2)
SUVLBW=              Formula 2
SUVBSA=              Formula 3
Changing SUVmax values normalised to BW at diff erent body weights 
Image 1: Impact of body weight to SUVmax at diff erent sphere volumes normalised to BW, when body weight is altered by 10% from the initial 
weight of the NEMA phantom (10 kg). The curves illustrate SUVmax value fl uctuations at diff erent body weights.
 0.5 ml 1.13 ml 2.5 ml 5.02 ml 11.01 ml 23.41 ml
10 kg 2.67 4.49 6.07 6.05 5.75 5.59
11 kg 2.94 4.94 6.67 6.66 6.33 6.15
12 kg 3.2 5.39 7.28 7.26 6.91 6.7
13 kg 3.47 5.84 7.89 7.87 7.48 7.26
14 kg 3.74 6.29 8.49 8.47 8.06 7.82
28 Medical Imaging and Radiotherapy Journal (MITRJ) 37 (1)
Image 2: Impact of body weight to SUVmean at diff erent sphere volumes normalised to BW, when body weight is altered by 10% from the initial 
weight of the NEMA phantom (10 kg). The curves illustrate the SUVmean value fl uctuations at diff erent body weights.
Image 3: Impact of body weight to SUVmax at diff erent sphere volumes normalised to BSA, when body weight is altered by 10% from the initial 
weight of the NEMA phantom (10 kg). The curves illustrate the SUVmax value fl uctuations at diff erent body weights.
Changing SUVmean values normalised to BW at diff erent body weights
Altered SUVmax values normalised to BSA at diff erent body weights
The erroneous entry of body weight can have a statistically 
signifi cant impact on SUVmax (p < 0.001) and SUVmean (p 
< 0.001) values.  Image 1 and 2 show a trend of changing 
SUVmax and SUVmean normalised to BW, if body weight is 
steadily increased by 10%. 
The normalisation of SUV to BSA at diff erent body weights 
showed a statistically signifi cant diff erence at SUVmax (p < 
0.001) and SUVmean (p < 0.001). Image 3 and 4 show SUVmax 
and SUVmean values at diff erent body weights.
The erroneous entry of application time in the protocol or a 
deviation between time on the applicator and time on PET/
CT scanner showed a statistically signifi cant diff erence at 
SUVmax (p < 0.001) and SUVmean (p < 0.001). Image 5 and 
6 show deviations between diff erent time points at SUVmax 
and SUVmean values.
The analysis of applied RP activity concentration showed a 
statistically signifi cant diff erence when a 3 or more % of lower 
or higher intravenously RP activity concentration is applied 
at SUVmax (p < 0.001) and at SUVmean (p < 0.001). SUVmax 
and SUVmean are increasing at a lower applied activity than 
recommended and decreasing at higher values. Image 7 and 
8 show diff erences in SUVmax and SUVmean values at a lower 
applied RP activity. 
DISCUSSION
Quantitative PET/CT is an important tool for diagnosis, 
prognosis and monitoring the response to oncological 
therapy. Many factors impact the quantitative assessment of 
SUV PET/CT. To understand these factors, I analysed the most 
common factors and compared them to the results of research 
conducted by other authors. 
Rep S. / Impact of acquisition parameters on the quantitative assessment of pet imaging – analysis of the nema phantom
 0.5 ml 1.13 ml 2.5 ml 5.02 ml 11.01 ml 23.41 ml
10 kg 2.2 3.04 3.57 4.35 4.49 4.47
11 kg 2.67 3.46 3.93 4.44 4.69 5.03
12 kg 2.79 3.65 4.28 4.85 5.11 5.48
13 kg 3.03 3.95 4.64 5.31 5.54 5.94
14 kg 3.25 4.41 5 5.72 5.96 6.4
 0.5 ml 1.13 ml 2.5 ml 5.02 ml 11.01 ml 23.41 ml
10 kg 0.41 0.7 0.94 0.94 0.89 0.87
11 kg 0.43 0.73 0.98 0.98 0.93 0.9
12 kg 0.45 0.75 1.02 1.02 0.97 0.94
13 kg 0.46 0.78 1.05 1.05 1 0.97
14 kg 0.48 0.81 1.09 1.09 1.03 1
Medical Imaging and Radiotherapy Journal (MITRJ) 37 (1) 29
Image 4: Impact of body weight to SUVmean at diff erent sphere volumes normalised to BSA, when body weight is altered by 10% from the 
initial weight of the NEMA phantom (10 kg). The curves illustrate the SUVmean value fl uctuations at diff erent body weights.
Image 5: SUVmax values at diff erent sphere volumes and time deviations as a consequence of the erroneous entry of application time or time 
discrepancies between applicator and PET/CT scanner. The curves illustrate the SUVmax value fl uctuations at time deviation.
Altered SUVmean values normalised to BSA at diff erent body weights
SUVmax values at time deviation
SUVmax and SUVmean are primarily normalised to BW. 
However, normalisation factors LBM and BSA are also used. 
Weber et al. (3), Lammertsma et al. (17), Young et al. (18) 
and Boellaard et al. (19) used analyses and determined that 
SUVmax and SUVmean normalised to BSA could have been 
more appropriate in examinations, particularly when patients 
lose weight during therapy. I conducted this research on the 
NEMA body phantom and compared SUVmax and SUVmean 
normalised to BW, LBM and BSA, and came to the conclusion 
that diff erences in body weight can have a statistically 
signifi cant impact on SUVmax and SUVmean when they are 
normalised to BW and BSA. The results of other authors also 
show the impact of lost body weight during therapy to SUV. 
The most appropriate method for the normalisation of SUV 
is still the subject of discussion and thus needs to be unifi ed 
with multi-centre trials (3, 8, 9, 17-19).
Time is an important parameter that impacts the quantitative 
assessment of SUV. The correct calculation of SUV depends on 
a precise cross-calibration between the PET/CT machine and 
the activity/dose meter (calibrator) that is used for measuring 
the activity concentration of applied RP for the patient. A 
common problem may occur as the result of an erroneous time 
Rep S. / Impact of acquisition parameters on the quantitative assessment of pet imaging – analysis of the nema phantom
 0.5 ml 1.13 ml 2.5 ml 5.02 ml 11.01 ml 23.41 ml
10 kg 0.37 0.47 0.55 0.63 0.66 0.71
11 kg 0.38 0.49 0.58 0.66 0.69 0.73
12 kg 0.39 0.51 0.6 0.69 0.72 0.77
13 kg 0.4 0.53 0.62 0.71 0.74 0.8
14 kg 0.42 0.54 0.64 0.73 0.76 0.82
 0 min 1 min 3 min 7 min 15 min 30 min 45 min 60 min
0.5 ml 2.67 2.69 2.72 2.79 2.93 3.23 3.55 3.9
1.13 ml 4.49 4.52 4.58 4.7 4.94 5.43 5.97 6.56
2.5 ml 6.07 6.11 6.18 6.34 6.67 7.33 8.06 8.68
5.02 ml 6.05 6.09 6.17 6.33 6.65 7.31 8.04 8.84
11.01 ml 5.75 5.79 5.86 6.01 6.33 6.95 7.65 8.4
23.41 ml 5.59 5.62 5.69 5.84 6.14 6.75 7.42 8.16
30 Medical Imaging and Radiotherapy Journal (MITRJ) 37 (1)
Image 6: SUVmean values at diff erent sphere volumes and time deviations as a consequence of the erroneous entry of application time or time 
discrepancies between applicator and PET/CT scanner. The curves illustrate the SUVmax value fl uctuations at time deviation.
Image 7: SUVmax values of diff erent sphere volumes and diff erent radiopharmaceutical activities. The curves illustrate the SUVmax value 
fl uctuations at diff erent radiopharmaceutical activities.
SUVmean values at time deviation 
SUVmax values in diff erent applied activities
Rep S. / Impact of acquisition parameters on the quantitative assessment of pet imaging – analysis of the nema phantom
synchronisation on a PET/CT scanner and time on the activity/
dose meter (calibrator) or read-out computer. RP is prepared 
for a patient and defi ned for a specifi c time unit, which is 
usually not entirely the same as the actual application time. It 
is therefore necessary to use the right corrections of physical 
decay of RP. This means that the RP activity for application must 
be defi ned on the basis of RP preparation and application, and 
the time when PET/CT imaging begins. The obtained analysis 
results confi rm the impact of time on SUV. An error can be the 
result of erroneous time synchronisation between PET/CT and 
dose meter (calibrator) or a consequence of erroneous time 
entry in the imaging protocol. The collected results match the 
results of other studies (3, 12, 13, 19). 
The net activity/dose of prepared RP that is administered to 
a patient must be measured precisely and applied in whole. 
It must be ensured that the remaining activity after the RP 
application in the injector is minimised to 1%. The remaining 
activity in the injector can be measured after use. It is lower than 
3% of the defi ned dose in most cases (95%). It is necessary to 
know the exact net activity/dose prepared for a patient. In 5% 
of all cases, the remaining activity in the injector accounts for 
10% (19). This is mostly due to a very high specifi c RP activity 
(RP activity on a total amount or mass MBq/ml), i.e. soon after 
production). To avoid this, the empty volume in the injector 
and the application process must be taken into account. The 
aforementioned problems can arise when RP is prepared and 
 0 min 1 min 3 min 7 min 15 min 30 min 45 min 60 min
0.5 ml 2.2 2.24 2.28 2.34 2.61 2.87 3.16 3.47
1.13 ml 3.04 3.06 3.1 3.18 3.34 3.67 4.04 4.44
2.5 ml 3.57 3.59 3.64 3.73 3.92 4.31 4.74 5.21
5.02 ml 4.08 4.11 4.16 4.22 4.49 4.93 5.42 5.96
11.01 ml 4.26 4.29 4.34 4.45 4.68 5.15 5.66 6.22
23.41 ml 4.56 4.61 4.71 4.76 5.01 5.53 6.06 6.66
 0.5 1.13 2.5 5.02 11.01 23.41
60 MBq 2.67 4.49 6.07 6.05 5.75 5.59
57 MBq 2.81 4.74 6.39 6.37 6.06 5.88
54 MBq 2.97 4.99 6.74 6.72 6.39 6.21
51 MBq 3.14 5.29 7.14 7.12 6.77 6.57
48 MBq 3.34 5.62 7.58 7.57 7.19 6.98
Medical Imaging and Radiotherapy Journal (MITRJ) 37 (1) 31
Image 8: SUVmean values of diff erent sphere volumes and diff erent radiopharmaceutical activities. The curves illustrate the SUVmean value 
fl uctuations at diff erent radiopharmaceutical activities.
SUVmean in diff erent applied activities
Rep S. / Impact of acquisition parameters on the quantitative assessment of pet imaging – analysis of the nema phantom
applied manually. Using an automatic applicator can eliminate 
these issues in most cases. If RP is applied paravenously, the 
quantitative assessment of SUV is not objective. It is, however, 
still possible and a visual assessment of the PET/CT image 
is facilitated. The performed analysis confi rmed that an 
incomplete application or inappropriately prepared activity 
can have a signifi cant impact on the quantitative value of 
SUVmax and SUVmean, and matched the results published by 
other authors (3, 19).
PET/CT work includes doctors, medical physicists, registered 
nurses and graduate radiographers. Radiographers are 
responsible for a correctly performed examination, which 
also includes the correct entry of data in imaging protocol 
(RP applied activity, RP application time, patient’s weight and 
height) and RP application. 
Radiographers regularly conduct routine tests (daily and 
monthly quality control tests (QC)) on a PET/CT machine and 
applicator, and annual tests together with medical physicists.  
CONCLUSION
Quantitative assessment of SUVmax and SUVmean is an 
important tool in oncology imaging assessment. The analyses 
showed that parameters, which include incorrect time 
between the RP application and imaging time, incomplete 
application or incorrectly measured RP activity as well as 
incorrect or incorrectly entered body weight of the patient 
into imaging protocol, can have a signifi cant impact and alter 
the quantitative assessment of SUVmax and SUVmean. Since 
a radiographer bears a great responsibility when performing 
imaging, they must be aware of potential errors that can 
lead to an incorrect quantitative assessment of SUVmax and 
SUVmean in the PET/CT examination.
REFERENCES
1. Juweid ME, Stroobants S, Hoekstra OS, et al. Use of 
positron emission tomography for response assessment 
of lymphoma: consensus of the Imaging Subcommittee 
of International Harmonization Project in Lymphoma. J 
Clin Oncol. 2007 Feb 10;25(5):571–578.
2. Larson SM, Schwartz LH. 18F-FDG PET as a candidate for 
‘‘qualifi ed biomarker’’: functional assessment of treatment 
response in oncology. J Nucl Med. 2006 Jun;47(6):901–
903.
3. Weber WA. Use of PET for monitoring cancer therapy and 
for predicting outcome. J Nucl Med. 2005 Jun;46(6):983–
995. 
4. Cheson BD, Pfi stner B, Juweid ME, et al. Revised response 
criteria for malignant lymphoma. J Clin Oncol. 2007 Feb 
10;25(5):579–586.
5. Weber WA. Chaperoning drug development with PET. J 
Nucl Med. 2006 May;47(5):735–737.
6. Sadato N, Tsuchida T, Nakamura S, et al. Non-invasive 
estimation of the net infl ux constant using the 
standardized uptake value for quantifi cation of FDG 
uptake of tumours. Eur J Nucl Med. 1998 Jun;25(6):559–
564.
7. Graham MM, Peterson LM, Hayward RM. Comparison of 
simplifi ed quantitative analyses of FDG uptake. Nucl Med 
Biol. 2000 Oct;27(7):647–655.
8. Stahl A, Ott K, Schwaiger M, Weber WA. Comparison of 
diff erent SUV-based methods for monitoring cytotoxic 
therapy with FDG PET. Eur J Nucl Med Mol Imaging. 2004 
Nov;31(11):1471–1479.
 0.5 1.13 2.5 5.02 11.01 23.41
60 MBq 2.2 3.04 3.57 4.08 4.26 4.56
57 MBq 2.5 3.2 3.79 4.3 4.48 4.8
54 MBq 2.64 3.38 3.96 4.49 4.73 5.09
51 MBq 2.8 3.57 4.2 4.8 5.01 5.36
48 MBq 2.9 3.8 4.46 5.1 5.33 5.7
32 Medical Imaging and Radiotherapy Journal (MITRJ) 37 (1)
9. Kim CK, Gupta NC. Dependency of standardized uptake 
values of fl uorine-18 fl uorodeoxyglucose on body 
size: comparison of body surface area correction and 
lean body mass correction. Nucl Med Commun. 1996 
Oct;17(10):890–894.
10. Zasadny KR, Wahl RL. Standardized uptake values of 
normal tissues at PET with 2-[fl uorine-18]-fl uoro-2-deoxy-
D-glucose: variations with body weight and a method for 
correction. Radiology. 1993;189:847–850.
11. Sugawara Y, Zasadny KR, Neuhoff  AW, Wahl RL. 
Reevaluation of the standardized uptake value for FDG: 
variations with body weight and methods for correction. 
Radiology. 1999 Nov; 213(2):521–525. 
12. Geworski L, Knoop BO, de Wit M, Ivancevic V, Bares R, 
Munz DL. Multicenter comparison of calibration and 
cross calibration of PET scanners. J Nucl Med. 2002 
May(5);43:635–639.
13. Westerterp M, Pruim J, Oyen W, et al. Quantifi cation 
of FDG PET studies using standardised uptake values 
in multi-centre trials: eff ects of image reconstruction, 
resolution and ROI defi nition parameters. Eur J Nucl Med 
Mol Imaging. 2007 Mar;34(3):392–404.
14. Hoekstra CJ, Hoekstra OS, Stroobants SG, et al. Methods 
to monitor response to chemotherapy in non–small 
cell lung cancer with 18F-FDG PET. J Nucl Med. 2002 
Oct;43(10):1304–1309.
Rep S. / Impact of acquisition parameters on the quantitative assessment of pet imaging – analysis of the nema phantom
15. Tahari AK, Chien D, Azadi JR, Wahl RL. Optimum Lean 
Body Formulation for Correction of Standardized Uptake 
Value in PET Imaging. J Nucl Med.  2014 Sep;55(9):1481–
4. 
16. Du Bois D and Du Bois E. A formula to estimate the 
approximate surface area if height and weight be known. 
Arch Intern Med 1916;17:863–71.
17. Lammertsma AA, Hoekstra CJ, Giaccone G, Hoekstra OS. 
How should we analyse FDG PET studies for monitoring 
tumour response? Eur J Nucl Med Mol Imaging. 2006 
Jul;33(suppl 1):16–21.
18. Young H, Baum R, Cremerius U, et al. Measurement 
of clinical and subclinical tumour response using 
[18F]-fl uorodeoxyglucose and positron emission 
tomography: review and 1999 EORTC recommendations. 
European Organization for Research and Treatment of 
Cancer (EORTC) PET Study Group. Eur J Cancer. 1999 
Dec;35(13):1773–1782.
19. Boellaard R, Oyen WJ, Hoekstra CJ, et al. The Netherlands 
protocol for standardisation and quantifi cation of FDG 
whole body PET studies in multicentre trials. Eur J Nucl 
Med Mol Imaging. 2008 Dec;35(12):2320–2333.