A Review on Motion Correction Methods in Pet/Ct Images for Detection of Cancer Cells

Abstract

Positron Emission Tomography (PET) is an important cancer imaging tool, both for diagnosing and staging, as well as offering predictive information based on response. PET is a nuclear medicine imaging technique which produces a three-dimensional image of functional processes in the body. While PET is commonly used to detect the tumors, especially in breast, colon, lung and for lymphoma, as well in the last decade it is verified as considerably more accurate than Computed Tomography (CT) in the distinction between benign and malignant lesions. PET is not only more accurate than conventional imaging for the assessment of therapy response, but also it is useful to detect some viable tumor cells after treatment. However, motion is a source of artifacts in the medical imaging and results in reducing the quantitative and qualitative accuracy of the image. In general during the procedure of PET scanning, a few types of motion can occur that should be corrected and compensated. Different body motions are classified as brain motion, cardiac motion and respiratory motion. In this study, some of the most important motion correction and compensation methods using PET imaging system are compared.

Full text

68
SCIENTIFIC REVIEW
A review on motion correction methods...
A REVIEW ON MOTION CORRECTION METHODS
IN PET/CT IMAGES FOR DETECTION OF CANCER CELLS
F. Nayyeri
Department of Computer Science
Faculty of Information Science and Technology
University Kebangsaan Malaysia
Summary. Positron Emission Tomography (PET) is an important cancer imag-
ing tool, both for diagnosing and staging, as well as offering predictive information
based on response. PET is a nuclear medicine imaging technique which produces
a three-dimensional image of functional processes in the body. While PET is com-
monly used to detect the tumors, especially in breast, colon, lung and for lymphoma,
as well in the last decade it is verifi ed as considerably more accurate than Computed
Tomography (CT) in the distinction between benign and malignant lesions. PET is
not only more accurate than conventional imaging for the assessment of therapy
response, but also it is useful to detect some viable tumor cells after treatment. How-
ever, motion is a source of artifacts in the medical imaging and results in reducing
the quantitative and qualitative accuracy of the image. In general during the proce-
dure of PET scanning, a few types of motion can occur that should be corrected and
compensated. Different body motions are classifi ed as brain motion, cardiac motion
and respiratory motion. In this study, some of the most important motion correction
and compensation methods using PET imaging system are compared.
Key words: positron emission tomography, computed tomography, motion correction,
motion compensation
INTRODUCTION
To ensure optimal treatment, an accurate and reliable staging modality
is essential (Takeuchi, Khiewvan et al. 2014). The conventional imag-
ing techniques of Magnetic Resonance Imaging (MRI) and Computed
Tomography (CT) with high spatial and contrast resolution are commonly used but
they still have expass some diffi culty to localize, visualize, and evaluate abnormal re-
10.1515/amb-2015-0020
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

Acta Medica Bulgarica, Vol. XLII, 2015, № 2 69
gions. Convincing evidences from several sources propose that although CT screen-
ing is the best method to detect cancer cells in its earliest stage, if a lymph node is
not enlarged or there is a spot in the tissue with the same density as the regular
tissue, or there is other spot like a benign cyst, CT scans have limitations in detect-
ing cancer. Both methods are potent for size and shape defi ning, but they can not
estimate if something is metabolically active or not (Saha 2010). However, PET, as
a functional imaging method, assesses the characterization and measurement of
the biologic processes. PET images generally lack anatomic context and have lower
spatial resolution than CT and MRI, but provide quantitative information about dis-
eases and structures (Bagci, Udupa et al. 2013).
PET scan is a potent method in determining whether some lesion is meta-
bolically active or not, but it generally can not determine precisely the size of the
lesion. Recently, combining unique strengths of functional and anatomical imaging
modalities, PET/CT has received much attention. PET/CT is the combination of PET
and CT imaging techniques within a single machine. The individual PET and CT
scans are taken concurrently while the patient remains in place, and can be pre-
sented separately or as a single, overlapping, “fused” image (Takeuchi, Khiewvan
et al. 2014).
In this context, patient’s movement becomes a major factor that degrades the
image quality. CT images are acquired within a few seconds, whereas PET images
are obtained over several minutes for each axial fi eld of view. Thus it is important to
take movements into account and respectively correct PET data. As a result, it will
considerably benefi t PET-based diagnosis and will make research in that fi eld more
reliable and accurate.
GLOBAL CANCER BURDEN IN 2012
The International Agency for Research on Cancer (IARC), which is specialized
cancer agency of the World Health Organization, releases the latest data on can-
cer incidence, mortality, and prevalence worldwide. According to GLOBOCAN Proj-
ect 2012 of the IARC, an estimated 14.1 million new cancer cases and 8.2 million
cancer-related deaths occurred in 2012, compared to 12.7 million and 7.6 million,
respectively, in 2008. Prevalence estimated for 2012 showed that there were 32.6
million people (over the age of 15 years) alive who had had a cancer diagnosed in
the previous fi ve years (Ferlay, Soerjomataram et al. 2014). Table 1 shows the most
common cancer types.
The most commonly diagnosed cancers worldwide were those of the lung (1.8
million, 13.0% of the total), breast (1.7 million, 11.9%), and colorectum (1.4 million,
9.7%). The most common causes of cancer death were cancers of the lung (1.6
million, 19.4% of the total), liver (0.8 million, 9.1%), and stomach (0.7 million, 8.8%)
(Ferlay, Soerjomataram et al. 2014).
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

70 A review on motion correction methods...
Table 1. The most common cancers in both sexes (GLOBOCAN 2012)
common LOBOCAN
Male Female
Cancer ASR*
(per 100,000) Cancer ASR
(per 100,000)
Incidence Mortality Incidence Mortality
Lung 34.9 30.9 Breast 47.9 14.9
Prostate 31.2 8.6 Colon 17.6 9.2
Colon 21 10.5 Lung 16.7 14
Stomach 17.7 13.2 Cervix uteri 15.1 7.6
Liver 15.6 14.6 Stomach 9.2 7.3
Bladder 9.3 3.5 Corpus uteri 9.1 2.2
Oesophagus 9.1 7.9 Ovary 6.8 4.3
Non-Hodgkin lymphoma 6.1 3.2 Thyroid 6.6 0.8
Kidney 6 2.6 Liver 6.5 6.4
Lip, oral cavity 5.6 2.8 Non-Hodgkin lymphoma 4.8 2.4
ASR* (Age-Standardised Rate): a population distribution of a standard age structure.
PET/CT IN DIAGNOSTIC PROCESS
Conventional imaging modalities such as CT or MRI, which use only dimensional
criteria to detect nodal involvement, have poor discriminatory power in differentiating
benign from malignant nodal disease (sensitivity range between 60% and 83% and
specifi city range between 77% and 82%) (Ambrosini, Nicolini et al. 2012). In the past
decades, CT has been employed as the gold standard imaging modality for cancer
cell staging. In fact, CT can accurately determine tumor size, mediastinal and vascular
invasion and can also suggest lymph node involvement, when the nodal axial diameter
is greater than 1 cm. In the following fi gure a mass in the left lung is shown on the CT
image (Figure 1a), while the combined PET/CT (Figure 1b) image reveals the meta-
bolic activity of that mass, as well as its precise location in the lung. The fused image
can help to diagnose and stage the disease, and to tailor the treatment plan.
(a) CT image (b) PET/CT image
Fig. 1. A mass in the left lung shown on (a) a CT image and on (b) a combined PET/CT image
(http://www.californiaheartandlungsurgery.com)
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

Acta Medica Bulgarica, Vol. XLII, 2015, № 2 71
PET/CT was reported to have a higher diagnostic accuracy than either PET or CT
alone. In addition, for the assessment of tumor, the integrated PET/CT imaging has in-
creased the accuracy of tumor detection compared to PET alone. PET/CT may provide
valuable information for the assessment of nodal stations that may be missed by conven-
tional imaging. Nevertheless, a recent cost-effectiveness analysis showed that PET/CT
can be recommended from an economic point of view (Ambrosini, Nicolini et al. 2012).
PET/CT IN RADIATION THERAPY (RT)
Recent studies show that despite improvements in survival rates for many tu-
mors, the 5-year overall survival for lung cancer remains relatively poor (around
10%) (Saha 2010, Bagci, Udupa et al. 2013) mainly because lung cancer is often
well advanced at the time of diagnosis and treatment options are limited (Takeuchi,
Khiewvan et al. 2014). Although surgery is the therapy of choice in early-stage lung
cancer, RT plays a major role in patients who are either medically or technically inop-
erable and who are not candidates for surgery (Lee, Kupelian et al. 2012).
Accurate staging of Non-Small Cell Lung Cancer (NSCLC) is essential for ap-
propriate therapy selection. Although the defi nition of volumes on PET images alone
might be more problematic due to the poorer resolution and higher noise levels,
when combined with structural imaging, such as CT, PET provides the best available
information on tumor extent.
Pre-therapy Post--therapy
Fig. 2. Post-radiation therapy with 18F-fl uoro-2-deoxyglucose positron emission tomography integrated
with computed tomography (18F-FDG PET/CT) shows signifi cant decrease in tumor volume.
Data courtesy of M. D. Anderson, Cancer Center, Orlando, FL, USA (Lee, Kupelian et al. 2012)
In the assessment of cancer, and more recently in infection diseases diagnosing,
fl uorodeoxyglucose (FDG) is the most commonly used PET radiotracer that enters tu-
mor cells provided by their increased glucose metabolism. The most metabolically ac-
tive tissues have the greatest need for sugar from the bloodstream which is measured
objectively as Standard Uptake Value (SUV). This is a number where higher number
means a higher metabolic rate which can be due to abnormal body activities. In other
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

72 A review on motion correction methods...
words, increased FDG uptake in metastatic nodes, which may not be signifi cantly
enlarged by CT criteria, enhances the sensitivity of PET/CT. In fact PET/CT should be
used for RT planning in NSCLC management because it displays tumor extent more
accurately than CT alone. The National Comprehensive Cancer Network guideline on
NSCLC states that PET/CT should be performed preferably 4 weeks before treatment
(Takeuchi, Khiewvan et al. 2014). However, sequential PET/CT imaging during RT
has been shown to be benefi cial for accurate evaluation of therapy response. Figure 2
shows FDG PET/CT before and after RT (Lee, Kupelian et al. 2012).
TYPE OF MOTION
In general during the procedure of PET scanning, a few types of motion can
occur. These different body motions are categorized as brain motion, cardiac motion
and respiratory motion. Each type needs its own specifi c approach to correct the
obtained images.
Brain PET Imaging
Unlike cardiac- and respiratory-related motions, patient movements in brain
imaging are assumed to be of rigid nature. Because a typical PET brain imaging
session can last hours, it is not reasonable to expect a patient to remain motion-
less during the whole time (Rahmim, Rousset et al. 2007). A number of head re-
straints are nowadays common, such as thermoplastic masks or neoprene caps
that lower the amount of motion but do not eliminate it (Rahmim, Rousset et al.
2007). Even with head restraints, typical translations in the range of 5 to 20 mm
and rotations of 1 to 4 degrees are observed, depending on the type of mask and
the scan duration.
Motion Due To The Cardiac Cycle
Although a spatial resolution of less than 5 mm is possible with current genera-
tion PET scanners, the base of the heart moves 9 to 14 mm toward the apex (Rah-
mim, Rousset et al. 2007). Compared to the primary resolution of today’s scanners,
cardiac motion can therefore result in signifi cantly blurred images. The most com-
mon approach to correct the cardiac cycle motions is gating the data into frames,
where each frame represents a particular cardiac phase. Typically, the cardiac cycle
is divided into 50- to 100-millisecond time frames, and an acquisition ranging from
5 to 60 minutes is usually acquired. Most commonly, the obtained cardiac-gated da-
tasets (i.e. cardiac frames) are independently reconstructed. This approach is suc-
cessful in removing nearly the cardiac-motion blurring of the images. However, it
usually produces images that are much noisier than a reconstruction of the ungated
data due to less statistics in each gated dataset in comparison with the entire dataset
(Rahmim, Rousset et al. 2007).
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

Acta Medica Bulgarica, Vol. XLII, 2015, № 2 73
Motion Caused By The Respiratory Cycle
Respiratory motion is a source of artifacts in thoracic and abdominal organs re-
sulting in a decrease in the quantitative and qualitative accuracy of the image. It has
been shown that during a quiet breathing in the spine position the diaphragm can
move by as much as 20mm and thus the liver can move by an average of 11 mm. It
is also well known that breathing induces rotational and/or translational movements
in thoracic and abdominal organs to varying extents. For example, upper areas of
the lungs are less subject to motion compared with the lower parts (Pépin, Daouk et
al. 2014). Therefore, for cancers located in the thorax or abdomen, the organ motion
caused by respiration remains a limiting factor in diagnostic imaging. The simplest
approach to respiratory motion issue is the “breath-holding”, but this limits acquisi-
tion time to typically less than 30 s, which is inadequate for PET imaging. Another
technique, called respiratory gating, is based on gated acquisition and synchroniza-
tion between the respiratory signal and PET acquisition. Respiratory gating involves
only acquired imaging data during a limited window (e.g. end-expiration), based on
a simple respiratory signal. However, this signifi cantly increases acquisition time.
MOTION TRACKING
An alternative solution for the motion correction is a motion tracking. In this tech-
nique, the respiratory signal is usually produced by an external sensor that tracks a
physiological characteristic related to the organ’s movement. As it can be diffi cult to
image the motion of interest directly during the procedure, markers are often implanted
into the region of interest and are tracked using an imaging device such as a camera
(McClelland, Hawkes et al. 2013). In this case the implantation can be invasive and
motion information is only available at the marker(s) and not for the whole region of
interest. Table 2 shows a summary of different types of motion correction techniques
and also shows the advantages and disadvantages of each technique.
Table 2. Summary of different motion correction methods
Year/Reference Motion Area Motion Correction Method Results
Li, Xie et al. 2010 Brain motion Developing the 3D volumetric image
registration (3DVIR) with sub-mm
accuracy to detect and correct subtle
misalignment in co-registered PET/
CT head images, using infrared motion
tracking camera and four different head
holding devices to reduce motion
Study on 53 patients found
misalignments in more than 80% of
image sets.
Fayad, Pan et al.
2010 Respiratory motion Reconstructing PET frames using the
one-pass list mode EM (OPL-EM)
algorithm and a realistic non-regular
patient-respiratory signal phantom.
Study on 2 nurbs-based cardiac-
torso (NCAT) phantoms and on
clinical data for 6 patients showed
improvement in the accuracy of the
PET respiratory corrected images.
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

74 A review on motion correction methods...
Chang, Chang et
al. 2010 Respiratory motion Proposing a joint motion and partial
volume effects (PVE) correction approach Study on 2 phantoms and 8
patients. found improvement in the
accuracy of PET quantifi cation by
simultaneously compensating for the
respiratory motion artifacts.
Marache-Francisco,
Lamare et al. 2010) Respiratory motion Evaluating two types of motion correction
techniques: averaging the co-registered
gated reconstructed PET images and
integrating the motion fi elds during the
iterative reconstruction process.
Study on 7 data series found clearly
visible tumor on corrected image
compared to hardly seen and not
distinguishable tumor from the noise
on the uncorrected one.
Ambwani, Karl et
al. 2011 Cardiac motion Cardiac Shape Tracking with Adjustment
for Respiration (CSTAR) using CT images
for cardiac shape tracking through the
estimation of cardiac motion
Study on simulated cardiac PET/
CT data corresponding to the XCAT
phantom found qualitative and
quantitative improvement compared
with conventional PET reconstruction.
McQuaid, Lambrou
et al. 2011 Cardiac motion Proposing a statistical shape model
to describe the diaphragm shape and
motion (diaphragm matching between
PET and CT)
Study on 2 patients revealed
quantitative improvements in the
PET images once they had been
corrected for attenuation and motion.
Keller, Sibomana et
al. 2012 Brain motion Proposing of 3 objective motion
correction methods with corresponding
results on human FDG brain scans,
using Polaris Vicra (Northern Digital Inc.)
and optical motion tracking with markers
attached to the head
Study on 17 patients Pros: effective
in motion larger than 2 mm Cons:
The fi xation method of the markers is
the main source of errors
Anishchenko, Hui et
al. 2012 Brain motion Proposing a head pose estimation
system for motion correction, using
external motion tracker to detect the head
pose with a photo camera and two web
cameras
Study on 12 patients Cons: head
rotation should happen only in one
plane and relative to one axis (in
real conditions, head movement can
happen in any direction), this method
cannot detect head movements less
than the tracker precision. In this
situations new additional blurriness
will be introduced
Noonan, Howard et
al. 2012 Respiratory motion Developing a low cost, high accuracy
solution for tracking respiratory motion by
producing a respiratory signal phantom to
enable gating of PET list mode data
Study on a National Electrical
manufacturers Association (NEMA)
image phantom data improving
motion correction of whole body
imaging by PET and camera-based
coincidence image techniques
RESPIRATORY MOTION COMPENSATION IN PET/CT IMAGING
The respiratory motion in thoracic PET images may lead to misidentifi cation
of the lesion or overestimation of its size. To solve the problem of motion com-
pensation of PET data, integrated PET/CT scan system is performed where CT
scan is acquired prior to PET acquisition. If the CT scan is performed during the
normal breathing of the patient, the artifacts produced in the CT image are further
transmitted into the PET image due to attenuation correction using the corrupted
Continned from the tabl. 2
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

Acta Medica Bulgarica, Vol. XLII, 2015, № 2 75
CT image. For these reasons breath-holding is recommended technique in CT
acquisition. As the CT scan is completed in a few seconds, breath-holding is a
possible approach. But PET data acquisition contains the whole moving process
and when use a breath-hold CT scan is only used, a part of the PET data can
be corrected for attenuation (Bai and Brady 2011). Due to motion, especially
for lesions near the boundary between lung and liver, the intensity of a lesion is
reduced, its size is overestimated and it is seriously mislocalized. These effects
lead to inaccurate lung cancer diagnosing and to a staging, which depend on the
intensity, the extent, and the location of a lesion. For all mentioned reasons, to
yield accurate and clear images and to avoid misled diagnosis, respiratory mo-
tion in PET data must be corrected.
Therefore, to improve the level of confi dence in PET scan images an improve-
ment in respiratory motion artifacts is needed. One method to compensate these
motion artifacts is the respiratory gating. In this method PET image data are acquired
into discrete bins within each breathing cycle where the fi rst bin is triggered at the
user preset position in the breathing cycle. To minimize the lesion motion, the num-
ber of bins within the patient’s respiratory cycle is optimized.
Nehmeh et al. in 2002 (Nehmeh, Erdi et al. 2002) developed a gating technique
to account for respiratory motion which decreased the accuracy in PET imaging of
lung cancer. Respiratory gating reduced the activity smearing and improved the ac-
curacy in identifying the tumor. In this article clinical data for fi ve patients with lung
cancer showed the potential benefi ts of respiratory gating PET imaging.
In the study of Bettinardi et al. (Bettinardi, Picchio et al. 2010) the purpose is to
describe the degradation effects produced by respiratory organ and lesion motion on
PET/CT images and to defi ne the role of respiratory gated 4D-PET/CT techniques to
compensate for such effects.
He et al. (He, O’Keefe et al. 2010) proposed a new method for gating respira-
tory motion in a 3D PET scanner system. This non-invasive method needs no addi-
tional hardware device to utilize the geometric sensitivity of PET imaging. However,
being insensitive to non-axial motion is the drawback of this method. Since the organ
motion is in the “z” direction within the respiratory cycle, the proposed method by
simulating PET acquisition and respiratory motion is able to improve the image deg-
radation caused by the respiratory motion.
In 2011, a method for motion and attenuation correction is proposed in respi-
ratory gated PET images. According to experimental results this method is able to
correct effectively the motion and improve the PET image quality. In addition, this
method provides improved diagnostic information without increasing the acquisition
duration or radiation burden. However, the major drawback of this method is us-
ing a time-varying system matrix for handling the respiratory motion. This strategy
performs motion correction by a very large matrix. The dimensions of this matrix are
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

76 A review on motion correction methods...
determined by the number of voxels and the number of projection bins, thus calcu-
lating such a matrix carries a major challenge for computational resources (Bai and
Brady 2011).
As mentioned earlier, most of the artifact corrections are based on synchroni-
zation between the respiratory signal and PET acquisition. The respiratory signal is
usually produced by an external sensor that is responsible for tracking the physiolog-
ical characteristic related to breathing (Pépin, Daouk et al. 2014). Respiratory gating
is a technique for reconstructed PET images that excludes the motion by using time
or amplitude binning. Although this technique is normally applied in clinical images,
it is not able to completely correct the breathing motion because each gate can mix
several tissue positions. The most important possibility to solve this problem is PET
acquisition duration. The fi rst solution is increasing the time by either selecting PET
events from gated acquisitions or performing several PET acquisitions. Therefore,
enough counting statistics are to be obtained in the different gates after binning. The
second solution is not to increase the acquisition duration, which suggests taking
into account all counting statistics and integrating motion information before, during
or after the reconstruction process. In the Table 3 the literature about the respiratory
motion is summarized.
Table 3. Summary of different motion correction methods
Reference /Year Summary
Nehmeh, Erdi et al. 2002 Using the Respiratory gating technique in PET imaging to reduce activity smearing
and to improve accuracy in identifying the tumor.
Fredberg Persson, Nygaard et al. 2011 Comparison between 3 CT methods of three-dimensional CT (3D-CT), four-
dimensional CT (4D-CT) and breath-hold CT (BH-CT) scan for evaluating the
tumor size. Results: breath-holding causes size of tumor to be presented smaller
in CT methods.
He, O’Keefe et al. 2010 Proposed a new method by simulating PET acquisition and respiratory motion to
improve the image degradation caused by respiratory motion. Drawback: being
insensitive to non-axial motion
Bettinardi, Picchio et al. 2010 Two solutions for the problem of respiratory gating to completely correct the
breathing motion:
1) Increasing the PET acquisition time to obtain enough counting statistics.
2) Take all counting statistics and integrate motion information before, during or
after the reconstruction process.
Bai and Brady 2011 Proposed a method to correct the motion in respiratory gated PET images and to
yield an appropriate distribution of the radioactivity concentration. Drawback: using
a very large time-varying system matrix for handling the respiratory motion and its
calculating is a major challenge for computational resources.
Pépin, Daouk et al. 2014 Aim: to defi ne the role of respiratory gated 4D-PET/CT techniques to compensate
for degradation effects produced by respiratory motion on PET/CT images.
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

Acta Medica Bulgarica, Vol. XLII, 2015, № 2 77
CONCLUSIONS
PET imaging provides quantitative functional information on diseases but mo-
tion caused by organ movement makes images blurred. Therefore, motion correc-
tion is of great importance for improving the quantity and quality of the obtained
images. In this paper, we presented the state-of-the-art motion correction methods
of image reconstruction that are commonly used for PET/CT imaging, as well as the
recent advances in techniques applicable to PET and PET/CT images. In this review
we described different motion correction methods, listed and compared which pro-
vides researchers and clinicians with detailed information and details that are well
suited for any particular application.
REFERENCE
1. Ambrosini, V. et al. PET/CT imaging in different types of lung cancer: an overview. – European
Journal of Radiology, 81, 2012, 5, 988-1001.
2. Ambwani, S. et al. Joint cardiac and respiratory motion correction and super-resolution reconstruc-
tion in coronary PET/CT. Biomedical Imaging: From Nano to Macro, IEEE International Symposium
on 2011, 2011, IEEE.
3. Anishchenko, S. et al. Quality Estimation Of Motion Correction For Pet Brain Images. – Southern
Federal University, 2012.
4. Bagci, U. et al. Joint segmentation of anatomical and functional images: Applications in quantifi ca-
tion of lesions from PET, PET-CT, MRI-PET, and MRI-PET-CT images. – Medical Image Analysis,
2013, 8, 929-945.
5. Bai, W. and M. Brady. Motion correction and attenuation correction for respiratory gated PET im-
ages. – Medical Imaging, 30, IEEE Transactions on 2011, 2, 351-365.
6. Bettinardi, V. et al. Detection and compensation of organ/lesion motion using 4D-PET/CT respiratory
gated acquisition techniques. – Radiotherapy and Oncology, 96, 2010, 3, 311-316.
7. Chang, G. et al. Joint correction of respiratory motion artifact and partial volume effect in lung/tho-
racic PET/CT imaging. – Medical Physics, 37, 2010, 12, 6221-6232.
8. Fayad, H. et al. A generic respiratory motion model for motion correction in PET/CT. Nuclear Sci-
ence Symposium Conference Record (NSS/MIC), 2010, IEEE 2010.
9. Ferlay, J. et al. GLOBOCAN 2012 v1. 0, Cancer Incidence and Mortality Worldwide: IARC Cancer-
Base № 11. Lyon, France: International Agency for Research on Cancer; 2013. – Last accessed
April 2014.
10. Fredberg Persson, G. et al. Artifacts in conventional computed tomography (CT) and free breathing
four-dimensional CT induce uncertainty in gross tumor volume determination. – International Jour-
nal of Radiation Oncology* Biology* Physics, 80, 2011, 5, 1573-1580.
11. He, J., et al. Motion compensation based on dynamic data in PET acquisition. Biomedical Engineer-
ing and Informatics (BMEI), 2010 3rd International Conference on 2010, IEEE.
12. Keller, S. H. et al. Methods for motion correction evaluation using 18F-FDG human brain scans on a
high-resolution PET scanner. – Journal of Nuclear Medicine 53, 2012, 3, 495-504.
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access

78 A review on motion correction methods...
13. Lee, P. et al. Current concepts in F18 FDG PET/CT-based radiation therapy planning for lung can-
cer. – Frontiers in Oncology, 2, 2012.
14. Li, G. et al. Correction of motion-induced misalignment in co-registered PET/CT and MRI (T1/T2/
FLAIR) head images for stereotactic radiosurgery. – Journal of Applied Clinical Medical Physics,
12, 2010, 1.
15. Marache-Francisco, S. et al. Impact of respiratory motion correction on the detection of small lesions
in whole-body PET imaging: a simulation study. Nuclear Science Symposium Conference Record
(NSS/MIC), IEEE, 2010.
16. McClelland, J. R. et al. Respiratory motion models: A review. – Medical Image Analysis, 17, 2013, 1, 19-42.
17. McQuaid, S. J. et al. A novel method for incorporating respiratory-matched attenuation correction
in the motion correction of cardiac PET–CT studies. – Physics in Medicine and Biology, 56, 2011,
10, 2903.
18. Nehmeh, S. A. et al. Effect of respiratory gating on quantifying PET images of lung cancer. – Journal
of Nuclear Medicine, 43, 2002, 7, 876-881.
19. Noonan, P. et al. Accurate markerless respiratory tracking for gated whole body PET using the
Microsoft Kinect. Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC),
IEEE, 2012.
20. Pépin, A. et al. Management of respiratory motion in PET/computed tomography: the state of the
art. – Nuclear Medicine Communications, 35, 2014, 2, 113.
21. Rahmim, A. et al. Strategies for motion tracking and correction in PET. – PET Clinics, 2, 2007, 2,
251-266.
22. Saha, G. B. Data Acquisition and Corrections. – Basics of PET Imaging, Springer, 2010: 41-69.
23. Takeuchi, S. et al. Impact of initial PET/CT staging in terms of clinical stage, management plan, and
prognosis in 592 patients with non-small-cell lung cancer. – European Journal of Nuclear Medicine
and Molecular Imaging, 41, 2014, 5, 906-914.
 Corresponding author:
F. Nayyeri
Department of Computer Science
Faculty of Information Science and Technology
University Kebangsaan Malaysia
43600 UKM Bangi, Selangor, Malaysia
tel.: (+60)14 701 5371
e-mail: fereshteh.nayyeri@gmail.com
- 10.1515/amb-2015-0020
Downloaded from PubFactory at 08/03/2016 11:53:54AM
via free access