Determining parameters for respiration-gated radiotherapy
S. S. Vedam
Departments of Biomedical Engineering and Radiation Oncology, Medical College of Virginia Hospitals
at Virginia Commonwealth University, Richmond, Virginia 23298
P. J. Keall,a) V. R. Kini, and R. Mohan
Department of Radiation Oncology, Medical College of Virginia Hospitals at Virginia Commonwealth
University, Richmond, Virginia 23298
!Received 11 September 2000; accepted for publication 9 July 2001"
Respiration-gated radiotherapy for tumor sites affected by respiratory motion will potentially improve radiotherapy outcomes by allowing reduced treatment margins leading to decreased complication rates and/or increased tumor control. Furthermore, for intensity-modulated radiotherapy
!IMRT", respiratory gating will minimize the hot and cold spot artifacts in dose distributions that
may occur as a result of interplay between respiratory motion and leaf motion. Most implementations of respiration gating rely on the real time knowledge of the relative position of the internal
anatomy being treated with respect to that of an external marker. A method to determine the
amplitude of motion and account for any difference in phase between the internal tumor motion and
external marker motion has been developed. Treating patients using gating requires several clinical
decisions, such as whether to gate during inhale or exhale, whether to use phase or amplitude
tracking of the respiratory signal, and by how much the intrafraction tumor motion can be decreased
at the cost of increased delivery time. These parameters may change from patient to patient. A
method has been developed to provide the data necessary to make decisions as to the CTV to PTV
margins to apply to a gated treatment plan. © 2001 American Association of Physicists in Medicine. #DOI: 10.1118/1.1406524$
Key words: respiration-gated radiotherapy, phase difference, duty cycle
I. INTRODUCTION
Respiration gating has been successfully applied to patient
treatments, first in Japan1–3 and more recently in the United
States.4,5 In this technique, the imaging and treatment devices are periodically turned on and off, in phase with the
patient breathing pattern, in order to restrict the range of
positions of the tumor and internal anatomy during imaging
and radiation delivery. This virtual restriction of position allows the internal margin component of the CTV to PTV6,7
expansion to be reduced, potentially reducing normal tissue
toxicity and/or allowing dose escalation and hence increased
tumor control.8 –10 The respiration signal can be monitored in
several ways by employing a video camera, spirometer, fluoroscopy, or strain gauge based systems.
For IMRT, gating has the added advantage of possibly
reducing unplanned hot and cold spots in the delivered dose
distribution. This problem occurs due to the interplay between the motion of the tumor and critical structures and the
motion of the MLC leaves used to deliver IMRT. Such motion may produce hot and cold spot artifacts in dose
distributions.11 These artifacts are similar to the motion artifacts seen in computed tomography !CT" images of moving
objects.12–16
Reduction of respiratory motion can be achieved using
either breathhold techniques4,17–22 or respiration gating
techniques.1–3,5,15,23–27 This study concentrates on respiration
gating as not all patients can maintain a breathhold for a
useful length of time,4,15,23 and respiration gating is less in2139
Med. Phys. 28 „10…, October 2001
terventional than breath hold. One could also consider an
intermediate approach in which the patient may hold his
breath for a variable duration depending upon his or her
comfort level. Such an approach will be more efficient than
respiratory gating and more comfortable for the patient than
the breath-hold techniques. This approach may be important
for IMRT due to its long treatment times. The efficiency of
any of the techniques to reduce the effects of internal motion
improves with patient breathing reproducibility.25
To determine the optimal gating parameters, two binary
decisions and one continuous decision need to be made. Gating can be performed at any stage of the breathing cycle,
however at inhale and exhale the mobility of internal
anatomy is at its minimum !consider the gradient of the cosine !%" curve at %!0 and &", and therefore gating should be
performed about the minimum or maximum point of motion.
Thus, whether to gate at inhale or exhale is our first decision.
The second decision is whether to track the respiratory signal
using the amplitude or the phase of the signal. Amplitude
gating is when the imaging and treatment are triggered when
the breathing trace is at a certain position. In phase gating,
the imaging and treatment are triggered when the calculated
breathing phase is at a certain angular phase, e.g., 30° before
full exhalation. Schematic diagrams of amplitude and phase
gating are shown in Fig. 1. Amplitude gating has the advantage that it is related to the position of the target, which is of
course what we want to irradiate during radiotherapy. Neither method gives useful results when the breathing is not
reproducible from one cycle to the next.
0094-2405Õ2001Õ28„10…Õ2139Õ8Õ$18.00
© 2001 Am. Assoc. Phys. Med.
2139
2140
Vedam et al.: Respiration-gated radiotherapy
2140
FIG. 1. Schematic diagrams of !a" amplitude gating where the imaging/
treatment device is triggered according
to the position of the signal. The
imaging/treatment is triggered when
the position of the signal is between
the two horizontal lines !the position
of which is set by the user". !b" Phase
gating where the imaging/treatment
device is triggered according to the
phase of the signal. The imaging/
treatment is triggered when the phase
of the signal is between the two phase
limits !as seen on the phase trace at the
left of the strip chart".
The third decision is the cost/benefit consideration of how
long we wish to extend the delivery time for the sake of
decreased tumor motion. The ratio of beam-on time to treatment time is termed the duty cycle. The range of tumor position decreases with duty cycle leading to smaller treatment
margins. However, the radiation delivery time increases. As
the treatment time increases, the patient throughput decreases, and also the chance of patient motion due to discomfort increases. Any patient motion due to patients adjusting
their position negates the potential gains of respiration gating.
The aim of this work is to demonstrate a method to determine the optimal parameters for respiration-gated imaging
and therapy on a patient-by-patient basis. These parameters
are whether to gate on inhale or exhale, use amplitude or
phase tracking, and the optimal duty cycle from which the
margins used for planning are to be defined.
II. METHOD AND MATERIALS
The methods presented here are general to all potential
gating patients. However, due to the difficulty of segmenting
and tracking the tumor with time, this approach is currently
applicable only to tumors that move with the same magnitude and phase as the diaphragm, which include liver
tumors28 and lung tumors in the vicinity of the diaphragm.
To demonstrate these methods, they have been applied to the
data from two elderly male patients with nonsmall cell lung
cancer in the right lower lobe of the lung attached to the
diaphragm. These patients are referred to as patient A and
Medical Physics, Vol. 28, No. 10, October 2001
patient B. Patient A breathed with more diaphragm motion
and patient B with less than the average of those studies
reviewed by Langen and Jones.29
A. Gating equipment and setup
The gating equipment used was the Varian !Varian Medical Systems, Palo Alto, CA 94304" Real-time Position Management !RPM" system version 1.2.2, connected to the Kermath !Kermath Manufacturing Company, Richmond, VA"
simulator with fluoroscopic capability. The RPM system uses
an infrared illuminator to highlight infrared reflectors attached to a marker block placed on the patient’s abdomen or
chest, which is imaged using a CCD camera. The number of
expected pixels in each marker is specified. Very high contrast of the reflective markers is observed, which make image
segmentation easy. The markers are a calibrated distance
apart, and thus absolute motion in the plane perpendicular to
the camera can be obtained. The periodic motion of the
marker is tracked synchronously with the input fluoroscopy
signal from the simulator. The synchronous inputs allow a
time correlation of the internal motion seen on the fluoroscopy images and the external marker motion.
Figure 2 shows a volunteer in simulation position with the
infrared marker box placed on the abdomen. Inset is a picture
of the camera used to acquire the motion of the marker box.
B. Data acquisition
A commercially available software package !AVI Constructor Version 3.2. Copyright 1996 –99 Michael Caracena,
2141
Vedam et al.: Respiration-gated radiotherapy
2141
C. Determining phase shift between internal and
external anatomy
FIG. 2. Volunteer in simulation position showing the infrared reflective
markers placed midway between the umbilicus and the xyphoid process.
The camera with an infrared illuminator surrounding the lens is inset. !Note
that for patients the marker block is attached directly to the skin rather than
the clothing."
4412 Pali Way, Boulder, CO" was used to extract individual
images from the fluoroscopy video !avi" file. These images
were stored as independent indexed image files. Selected images obtained over half of a breathing cycle are shown in
Fig. 3. Information about the extent of motion of the diaphragm in the superior–inferior direction was obtained by
determining the vertical position of the diaphragm apex on
each image frame using an edge detection algorithm. The
apex coordinates were stored in a file along with the corresponding frame number of the image. This procedure was
repeated for all of the individual images. The automated
tracking was compared with manual tracking as performed
by three independent observers, and the results were highly
correlated.
During data acquisition, the marker trace was updated at
the rate of 30 frames per second while the fluoroscopy images were captured at 10 frames per second. The RPM software calculates both the position and phase of the marker
placed on the patient. The three marker trace points corresponding to each image were averaged. The obtained phase
and amplitude for each image were stored in a table along
with the diaphragm position.
FIG. 3. Selected images extracted from the video file from exhale to inhale
during normal breathing. Note the decreasing position of the diaphragm as
the patient inhales. The outlined GTV, attached to the diaphragm, can also
be seen to decrease in position.
Medical Physics, Vol. 28, No. 10, October 2001
Radiotherapy based on respiration gating using tracking
of external variables, such as chest/abdomen motion or airflow, relies on the existence of a consistent time correlation
between the external measurements and internal anatomy
motion. This correlation can either be in phase or out of
phase, depending on how each individual breathes, whether
he or she is a chest or abdomen breather, the position and
type of respiration monitoring device, and the anatomical site
being tracked. Thus, knowledge of the phase shift between
the internal anatomy to be treated and the external anatomy
being tracked is essential.
Reproducibility of the marker position upon the skin from
day to day is important, as phase shifts are observed between
different marker placements in the abdomen and skin. Also,
due to the variation in direction, amplitude, and phase correlation of the internal anatomy motion with the marker position, it is important that the internal anatomy for which the
motion is determined is the GTV, or else has a known motion
correlation to the GTV for situations in which the GTV itself
cannot be determined !as is often the case with fluoroscopic
examinations".
The phase difference between the position of the marker
and the internal anatomy was determined by aligning the
fluoroscopy tracking trace over the marker trace, until a
‘‘best’’ match was obtained !as judged by the user". The
number of frames over which the fluoroscopy trace had to be
aligned to obtain the best match was determined both in
terms of phase and time.
To investigate any systematic phase differences, the phase
difference was determined using the above-mentioned technique for a mechanical sinusoidal oscillator.30 The actual
phase difference for the device, in which the marker is rigidly attached to the moving ‘‘target,’’ is zero.
D. Determining gating type, position, and duty cycle
The range and standard deviation of positions within the
gated portions of the marker trace were determined as a function of duty cycle. The type of gating was chosen from
among four different options, namely, amplitude gating at
exhale, amplitude gating at inhale, phase gating at exhale,
and phase gating at inhale. The thresholds for amplitude gating were incremented in steps from the point of minimum
excursion of the marker trace to the point of maximum excursion, yielding range and standard deviation of motion values for duty cycles from 0 !' treatment time" to 1 !normal
treatment".
Gating thresholds !see Fig. 1" for phase gating were also
similarly varied starting from the minimum phase value to
the maximum phase value. For each gating threshold, the
program then extracted all data points that were within the
gating limits. Range of motion and standard deviation values
were calculated from points on the phase-shifted diaphragm
positions that occurred within the duty cycle. Plots of duty
cycle versus range of motion and standard deviation of po-
2142
Vedam et al.: Respiration-gated radiotherapy
2142
sition were then generated for amplitude and phase gating at
both inhale and exhale points.
The results of the plots of the amplitude and phase for
inhale and exhale as a function of duty cycle are needed to
determine which part !inhale/exhale" of the breathing cycle
to use, and which gating mode !amplitude or phase" to use.
As the goal is to minimize motion during treatment, the plot
with the smallest variation in position as a function of duty
cycle is optimal.
Once the inhale/exhale and gating mode decisions have
been made, the margins to add to the CTV to create the PTV
for the treatment need to be determined. Using the ICRU 62
!Ref. 7" formalism, the margins to add to the CTV are the
internal margin !IM" and the setup margin !SM". !The internal margin accounts for variations in size, shape, and position of the CTV in relation to anatomical reference points.
The setup margin accounts for uncertainties in patient-beam
positioning." A discussion of the margin to add to the GTV to
obtain the PTV is given in the Appendix.
Antolak and Rosen31 suggest adding a CTV to PTV margin of 1.65 times the standard deviation of the margin errors.
As the IM and SM are independent, the standard deviations
of these distributions !( IM and ( SM , respectively" should be
added in quadrature. Thus the CTV to PTV margin, M, is
FIG. 4. A flow chart illustrating the principal steps in the extraction of
parameters needed for gated radiation treatments.
2
2
" ( SM
.
M !1.65!( IM
III. RESULTS
Note that ( IM is composed of intrafraction motion, ( IMintra ,
due to respiration, and interfraction displacement, ( IMinter ,
due to the differences in the tumor position with respect to
the bony anatomy between fractions. ( IMinter also incorporates interfraction variations in the relation between the internal and external marker. The quantification of ( IMinter is
fundamental and crucial to the success of gating; however, it
is beyond the scope of this article and will be addressed in a
separate study. For this study we used the estimated ( IMinter
value of 0.2 cm.
As ( IMintra is dependent on the duty cycle, the treatment
margin is dependent on the duty cycle. The treatment margin
as a function of duty cycle was calculated using our fluoroscopic study data for ( IMintra !as described earlier" and Ekberg et al.’s32 data for ( SM !0.46 cm in the cranio-caudal
direction". Note also that the margin will vary in the craniocaudal and medial-lateral and anterior-posterior directions.
A flow chart of the parameter extraction methodology is
shown in Fig. 4.
Another factor that needs to be taken into account is the
intersession reproducibility of the respiration signal. The reproducibility needs to be established, and a separate margin
added for this source of error that is ignored in the current
work. We are currently working on !a" evaluating this error
using successive fluoroscopy images of patients taken over
different days and !b" developing audio prompts and visual
feedback methods to improve the day to day reproducibility
of respiration.
Medical Physics, Vol. 28, No. 10, October 2001
A. Determining phase shift between internal and
external anatomy
The systematic phase shift of the system was measured by
rigidly attaching a marker to a moving sphere. Simultaneously the marker was tracked while a fluoroscopy video
was acquired. The motion of the edge of the sphere in the
video was then determined. The results are shown in Fig. 5.
Figure 6 shows the marker motion trace and diaphragm
before !a" and after !b" the phase shift to account for the
FIG. 5. Marker motion and sphere edge motion trace. In this plot the marker
motion overlaps the sphere edge motion, showing the systematic error of the
phase shift measurement is zero. Note that the amplitude of the marker
motion is arbitrarily normalized.
2143
Vedam et al.: Respiration-gated radiotherapy
2143
FIG. 7. Range of motion as a function of duty cycle for phase and amplitude
gating at both inhale and exhale points for patients A !a" and patient B !b".
Note that the y-axes on the two plots have different scales.
FIG. 6. Marker motion and diaphragm edge motion before !a" and after !b"
the phase shift to account for the time difference between the diaphragm and
external skin motion for patient A. The marker motion and diaphragm edge
motion for patient B is also shown !c". Patient B exhibits no phase shift.
Note that the amplitude of the marker motion trace is arbitrarily normalized.
respiration difference between the diaphragm and external
skin motion. Patient A results reveal a phase shift of 42°,
corresponding to 12% of a breathing cycle, or a 0.3 s delay.
Figure 6!c" shows that patient B had a phase shift of 0°. The
phase shift should be accounted for during treatment by a
time delay in the triggering of the radiation ‘‘ON.’’
B. Determining gating type, position, and duty cycle
The range of motion as a function of duty cycle for phase
and amplitude gating at both inhale and exhale points for
Medical Physics, Vol. 28, No. 10, October 2001
patients A and B is shown in Fig. 7, and standard deviation
of motion as a function of duty cycle is shown in Fig. 8.
These graphs show an increasing trend for the standard deviation and range of motion with the increase in duty cycle.
Note that the amplitude of diaphragm motion for patient A
was nearly three times that of patient B. Another feature that
is evident from the figures is that, for a given value of duty
cycle, the standard deviation values at inhale are usually
greater than those at exhale. This in turn leads to the conclusion that gating during exhale is more reproducible than gating during inhale, which is consistent with results obtained
by others.2,4 However, this general conclusion may vary
from patient to patient, and should be assessed by the current
method on an individual basis.
The range of motion and standard deviation of position as
a function of duty cycle curves shown in Figs. 7 and 8 may
have been improved !become more convex and thus decrease
motion and increase duty cycle" if breathing coaching methods were used.
The treatment margins for a given duty cycle can be determined using either the range of motion results, or standard
deviation results. The range of motion analysis gives values
that can be used directly as the intrafraction composition of
the IM. However, the range of motion is only calculated
using the minimum and maximum points, and therefore is
very sensitive to outliers that may vary from day to day.
!This sensitivity is also seen by comparing the smoothness of
2144
Vedam et al.: Respiration-gated radiotherapy
2144
FIG. 9. Cranio-caudal CTV to PTV margins as a function of duty cycle using
the values of ( IM intra from Fig. 8, ( IM inter !2 mm" and Ekberg et al.’s !Ref.
32" data for ( SM !4.6 mm in the cranio-caudal direction". 1.65 times the
total uncertainty was used to calculate the margin !Ref. 31". The amplitude
gating at exhale data was used for the patient A calculation, and phase gating
at exhale data was used for the patient B calculation.
FIG. 8. Standard deviation of position as a function of duty cycle for phase
and amplitude gating at both inhale and exhale points for patient A !a" and
patient B !b". Note that the y-axes on the two plots have different scales.
the curves seen in Fig. 8 compared to those of Fig. 7." The
standard deviation of motion can be applied to the margin
calculation such that the probability of the tumor being inside the margin during the radiation delivery can be used.31,32
All data points are used in the standard deviation of motion
analysis, resulting in less sensitivity to outliers than the range
of motion values. A limitation of using the standard deviation
method is that the distribution assumed !normal in our case"
may not be the best method to characterize the spread of the
distribution of values as a function of duty cycle.
Using the amplitude !patient A" and phase !patient B"
gating at exhale standard deviation !( IMintra) values from Fig.
8, the CTV to PTV margin was calculated as described in
Sec. II. The CTV to PTV margin as a function of duty cycle
is shown in Fig. 9. For patient A, for duty cycle values from
0% to 40% the setup error dominates, from 40% to 60% the
setup error and internal motion both affect the margin, and
above 60% the margin is dominated by the internal motion.
For patient B, the small amplitude of motion means that the
CTV to PTV margin is dominated by the setup error for all
duty cycle values, and the use of margin would not result in
margin reduction.
The final decision is to determine which duty cycle is
optimal, by balancing the desired reduction in margins with
the corresponding increase in treatment time due to this reduction as the duty cycle is reduced. From the patient results
presented here for patient A, a 40% duty cycle would have
been selected as any further increase in duty cycle would
Medical Physics, Vol. 28, No. 10, October 2001
lead to a greater margin; however, any decrease in duty cycle
would not lead to a significantly smaller margin. For patient
B the use of gating would not result in decreased CTV to
PTV margins, though gating may still be useful in reducing
motion artifacts during imaging and !IMRT" delivery.
The results shown in Fig. 9 are dependent on the setup
error, and show that if the intrafraction motion is small compared to the setup error, then the CTV and PTV margin reduction is minimal. The CTV to PTV margin reduction as a
function of intrafraction motion and setup error is shown in
Fig. 10. Due to the quadrature summation, the margin will
tend to be dominated by the larger component, and thus decreasing setup error is important as well as intrafraction motion.
FIG. 10. CTV and PTV margins as a function of the standard deviations of
setup error and internal motion. Due to the quadrature summation, the larger
of the two quantities has the greatest influence on CTV to PTV margins.
2145
Vedam et al.: Respiration-gated radiotherapy
IV. DISCUSSION
A procedure has been developed for the determination of
optimal gated radiotherapy parameters. First, the phase shift
between motion of the internal anatomy and resultant motion
of an external marker is determined. Subsequently the range
of motion and standard deviation of position for inhale and
exhale, and phase and amplitude gating as a function of duty
cycle are determined. Other treatment errors, such as setup
error are then included to obtain the CTV to PTV margin as
a function of duty cycle. For this relation, the duty cycle
value that best compromises decreased margins with increased treatment time can be selected.
The choice of optimal parameters and procedures for
gated treatments should not depend solely on their numerical
values alone. There are other factors that may need to be
considered. For instance, it has been shown17 that inspiration
provides a larger lung volume that means that the fractional
treated lung volume is smaller, and also inspiration can provide a greater separation between the tumor and critical
structures, such as the spinal cord. Thus, even though the
point of exhale may provide the most stable portion of the
respiratory cycle, its choice for gating treatments should also
consider the increased fraction of lung tissue exposed to
higher doses.
An obvious limitation of the current method is that the
diaphragm and not the GTV motion is being tracked with
time. Thus we are limited to tumors in the lower lobe of the
lung or in the liver, which have a similar phase and amplitude of motion as the diaphragm. The segmentation of the
GTV from fluoroscopy is a subject of ongoing research.
Also in need of investigation is the quantification of the
daily variations between the tumor motion and the external
marker. The variations should be included into the calculation of CTV–PTV margins.
APPENDIX: OBTAINING THE CTV FROM THE GTV
The addition of the margin for suspected subclinical disease is, of course, a clinical decision. However, it is common
practice for lung tumor cases to add a margin to the GTV
that accounts for the suspected subclinical disease, the IM
and the SM, thus creating the PTV from the GTV without
explicitly specifying CTV. If the IM and SM are known, the
margin used for the CTV can be inferred.
Often the GTV to PTV margin is 2 cm in the craniocaudal direction and 1.5 cm in the anterior-posterior and
medial-lateral directions. As discussed in the text, the margin
to add for the IM and SM is 1.65 times the quadrature sum of
the respective standard deviations, thus
2
2
" ( SM
subclinical extension"1.65# !( IM
!2 cm !cranio-caudal"
and
2
2
" ( SM
subclinical extension"1.65# !( IM
!1.5 cm !medial–lateral and anterior–posterior".
Using margin values ( IMintra!0.7 cm in the cranio-caudal
Medical Physics, Vol. 28, No. 10, October 2001
2145
direction and 0.2 cm in the two orthogonal directions,
( IMinter!0.2 cm and ( SM!0.46 cm in the cranio-caudal direction and 0.39 in the two orthogonal directions, we obtain
the cranio-caudal subclinical extension !2$1.65
# !0.72 "0.462 !0.6 cm and medial–lateral/anterior–
posterior subclinical extension !1.5$1.65# !0.22 "0.392
!0.8 cm.
a
Author to whom correspondence should be addressed: P.O. Box 980058,
Richmond, VA 23298-0058. Electronic mail: [email protected]
1
K. Ohara, T. Okumura, T. Akisada, T. Inada, T. Mori, H. Yokota, and M.
B. Calguas, ‘‘Irradiation synchronized with respiration gate,’’ Int. J. Radiat. Oncol., Biol., Phys. 17, 853– 857 !1989".
2
T. Tada, K. Minakuchi, T. Fujioka, M. Sakurai, M. Koda, I. Kawase, T.
Nakajima, M. Nishioka, T. Tonai, and T. Kozuka, ‘‘Lung cancer: intermittent irradiation synchronized with respiratory motion-results of a pilot
study,’’ Radiology 207, 779–783 !1998".
3
S. Minohara, T. Kanai, M. Endo, K. Noda, and M. Kanazawa, ‘‘Respiratory gated irradiation system for heavy-ion radiotherapy,’’ Int. J. Radiat.
Oncol., Biol., Phys. 47!4", 1097–1103 !2000".
4
H. D. Kubo, P. M. Len, S. Minohara, and H. Mostafavi, ‘‘Breathingsynchronized radiotherapy program at the University of California Davis
Cancer Center,’’ Med. Phys. 27!2", 346 –353 !2000".
5
D. Scaperoth, ‘‘Why respiration gated radiotherapy?’’ presented at the
RPM Respiration gating training, Knoxville, TN, 2000.
6
ICRU, ICRU Report 50. Prescribing, Recording and Reporting Photon
Beam Therapy !International Commission on Radiation Units and Measurements, Bethesda, MD, 1993".
7
ICRU, ICRU Report 62. Prescribing, Recording and Reporting Photon
Beam Therapy (Supplement to ICRU Report 50) !International Commission on Radiation Units and Measurements, Bethesda, MD, 1999".
8
H. X. Zhang, W. B. Yin, and L. J. Zhang, ‘‘Curative radiotherapy of early
inoperable non-small cell lung cancer,’’ Radiother. Oncol. 14, 89–94
!1989".
9
D. E. Doseretz, M. J. Katin, and B. H. Blitzer, ‘‘Radiation therapy in the
management of medically inoperable carcinoma of the lung: results and
implications for future treatment strategies,’’ Int. J. Radiat. Oncol., Biol.,
Phys. 24, 3–9 !1992".
10
G. S. Silbey, ‘‘Radiotherapy for patients with medically inoperable stage
I nonsmall cell lung carcinoma,’’ Cancer !N.Y." 82, 433– 438 !1998".
11
C. X. Yu, D. A. Jaffray, and J. W. Wong, ‘‘The effects of intra-fraction
organ motion on the delivery of dynamic intensity modulation,’’ Phys.
Med. Biol. 43, 91–104 !1998".
12
L. A. Shepp, S. K. Hilal, and R. A. Schulz, ‘‘The tuning fork artifact in
computerized tomography,’’ Comput. Graph. Image Process. 10, 246 –
255 !1979".
13
J. R. Mayo, N. L. Müller, and R. M. Henkelman, ‘‘The double-fissure
sign: a motion artifact on thin-section CT scans,’’ Radiology 165, 580–
581 !1987".
14
R. D. Tarver, D. J. Conces, and J. D. Godwin, ‘‘Motion artifacts on CT
simulate bronchiectasis,’’ AJR, Am. J. Roentgenol. 151!6", 1117–1119
!1988".
15
C. J. Ritchie, J. Hseih, M. F. Gard, J. D. Godwin, Y. Kim, and C. R.
Crawford, ‘‘Predictive respiratory gating: A new method to reduce motion artifacts on CT scans,’’ Radiology 190!3", 847– 852 !1994".
16
S. Shimuzu, H. Shirato, K. Kagei, T. Nishioka, X. Bo, H. Dosaka-Akita,
S. Hashimoto, H. Aoyama, K. Tsuchiya, and K. Miyasaka, ‘‘Impact of
respiratory movement on the computed tomographic images of small
lung tumors in three-dimensional !3D" radiotherapy,’’ Int. J. Radiat. Oncol., Biol., Phys. 46!5", 1127–1133 !2000".
17
J. Hanley, M. M. Debois, D. Mah, G. S. Mageras, A. Raben, K. Rosenzweig, B. Mychalczak, L. H. Schwartz, P. J. Gloeggler, W. Lutz, C. C.
Ling, S. A. Leibel, Z. Fuks, and G. J. Kutcher, ‘‘Deep inspiration breathhold technique for lung tumors: The potential value of target immobilization reduced lung density in dose escalation,’’ Int. J. Radiat. Oncol.,
Biol., Phys. 45, 603– 611 !1999".
18
K. E. Rosenzweig, J. Hanley, D. Mah, G. Mageras, M. Hunt, S. Toner, C.
Burman, C. C. Ling, B. Mychalczak, Z. Fuks, and S. A. Leibel, ‘‘The
deep inspiration breath-hold technique in the treatment of inoperable nonsmall-cell lung cancer,’’ Int. J. Radiat. Oncol., Biol., Phys. 48!1", 81– 87
!2000".
2146
Vedam et al.: Respiration-gated radiotherapy
19
J. M. Balter, K. L. Lam, C. J. McGinn, T. S. Lawrence, and R. K. Ten
Haken, ‘‘Improvement of CT-based treatment-planning models of abdominal targets using static exhale imaging,’’ Int. J. Radiat. Oncol., Biol.,
Phys. 41!4", 939–943 !1998".
20
D. Mah, J. Hanley, K. E. Rosenzweig, E. Yorke, L. Braban, C. C. Ling, S.
A. Leibel, and G. Mageras, ‘‘Technical aspects of the deep inspiration
breath-hold technique in the treatment of thoracic cancer #in process citation$,’’ Int. J. Radiat. Oncol., Biol., Phys. 48!4", 1175–1185 !2000".
21
J. W. Wong, M. B. Sharpe, D. A. Jaffray, V. R. Kini, J. S. Robertson, J. S.
Stromberg, and A. A. Martinez, ‘‘The use of active breathing control
!ABC" to reduce margin for breathing motion,’’ Int. J. Radiat. Oncol.,
Biol., Phys. 44, 911–919 !1999".
22
J. S. Stromberg, M. B. Sharpe, L. H. Kim, V. R. Kini, D. A. Jaffray, A. A.
Martinez, and J. W. Wong, ‘‘Active breathing control !ABC" for
Hodgkin’s disease: reduction in normal tissue irradiation with deep inspiration and implications for treatment #in process citation$,’’ Int. J. Radiat.
Oncol., Biol., Phys. 48!3", 797– 806 !2000".
23
M. Mori, K. Murata, M. Takahashi, K. Shimoyama, T. Ota, R. Morita,
and T. Sakamoto, ‘‘Accurate contiguous sections without breath-holding
on chest CT: Value of respiratory gating and ultrafast CT,’’ AJR, Am. J.
Roentgenol. 162!5", 1057–1062 !1994".
24
H. D. Kubo and B. C. Hill, ‘‘Respiration gated radiotherapy treatment: a
technical study,’’ Phys. Med. Biol. 41, 83–91 !1996".
25
G. S. Mageras, ‘‘International strategies for reducing respiratory-induced
motion in external beam therapy,’’ in The Use of Computers in Radiation
Medical Physics, Vol. 28, No. 10, October 2001
2146
Therapy, edited by W. Schlegel and T. Bortfeld !Springer, Heidelberg,
2000", pp. 514 –516.
26
C. R. Ramsey, D. Scaperoth, D. Arwood, and A. L. Oliver, ‘‘Clinical
efficacy of respiratory gated conformal radiation therapy,’’ Med. Dosim.
24!2", 115–119 !1999".
27
H. Shirato, S. Shimizu, T. Kunieda, K. Kitamura, M. van Herk, K. Kagei,
T. Nishioka, S. Hashimoto, K. Fujita, H. Aoyama, K. Tsuchiya, K. Kubo,
and K. Miyasaka, ‘‘Physical aspects of real-time tumor-tracking system
for gated radiotherapy #in process citation$,’’ Int. J. Radiat. Oncol., Biol.,
Phys. 48!4", 1187–1195 !2000".
28
L. A. Dawson, K. Brock, D. Litzenberg, S. Kazanjian, C. J. McGinn, T. S.
Lawrence, and R. K. Ten Haken, ‘‘The reproducibility of organ position
using active breathing control !ABC" during liver radiotherapy,’’ Int. J.
Radiat. Oncol., Biol., Phys. 48!3S", 165 !2000".
29
K. M. Langen and D. T. Jones, ‘‘Organ motion and its management,’’ Int.
J. Radiat. Oncol., Biol., Phys. 50!1", 265–278 !2001".
30
P. J. Keall, V. Kini, S. S. Vedam, and R. Mohan, ‘‘Motion Adaptive X-ray
Therapy: A feasibility study,’’ Phys. Med. Biol. 46!1", 1–10 !2001".
31
J. A. Antolak and I. I. Rosen, ‘‘Planning target volumes for radiotherapy:
How much margin is needed?’’ Int. J. Radiat. Oncol., Biol., Phys. 44!5",
1165–1170 !1999".
32
L. Ekberg, O. Holmberg, L. Wittgren, G. Bjelkengren, and T. Landberg,
‘‘What margins should be added to the clinical target volume in radiotherapy treatment planning for lung cancer?’’ Radiother. Oncol. 48,
71–77 !1998".