4dct rt planning red 2006

7/31/2019 4dct Rt Planning Red 2006

1/12

doi:10.1016/j.ijrobp.2006.04.031

PHYSICS CONTRIBUTION

MID-VENTILATION CT SCAN CONSTRUCTION FROM FOUR-DIMENSIONALRESPIRATION-CORRELATED CT SCANS FOR RADIOTHERAPY PLANNING

OF LUNG CANCER PATIENTS

JOCHEM W. H. WOLTHAUS, M.SC., CHRISTOPH SCHNEIDER, PH.D., JAN-JAKOB SONKE, PH.D.,MARCEL VAN HERK, PH.D., JOS S. A. BELDERBOS, M.D.,

MADDALENA M. G. ROSSI, D.C.R.(R), R.T.T., JOOS V. LEBESQUE, M.D., PH.D.,AND EUGNE M. F. DAMEN, PH.D.

Department of Radiation Oncology, The Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital,Amsterdam, The Netherlands

Purpose: Four-dimensional (4D) respiration-correlated imaging techniques can be used to obtain (respiration)artifact-free computed tomography (CT) images of the thorax. Current radiotherapy planning systems, however,do not accommodate 4D-CT data. The purpose of this study was to develop a simple, new concept to incorporatepatient-specific motion information, using 4D-CT scans, in the radiotherapy planning process of lung cancerpatients to enable smaller error margins.Methods and Materials: A single CT scan was selected from the 4D-CT data set. This scan represented the tumorin its time-averaged position over the respiratory cycle (the mid-ventilation CT scan). To select the appropriateCT scan, two methods were used. First, the three-dimensional tumor motion was analyzed semiautomatically tocalculate the mean tumor position and the corresponding respiration phase. An alternative automated methodwas developed to select the correct CT scan using the diaphragm motion.Results: Owing to hysteresis, mid-ventilation selection using the three-dimensional tumor motion had a tumorposition accuracy (with respect to the mean tumor position) better than 1.1 1.1 mm for all three directions(inhalation and exhalation). The accuracy in the diaphragm motion method was better than 1.1 1.1 mm.Conventional free-breathing CT scanning had an accuracy better than 0 3.9 mm. The mid-ventilation concept

can result in an average irradiated volume reduction of 20% for tumors with a diameter of 40 mm.Conclusion: Tumor motion and the diaphragm motion method can be used to select the (artifact-free) mid-ventilation CT scan, enabling a significant reduction of the irradiated volume. 2006 Elsevier Inc.

Computed tomography, Four-dimensional, Respiration-correlated, Breathing, Treatment planning, Lung can-cer, Mid-ventilation.

INTRODUCTION

Treatment outcomes for nonsmall-cell lung cancer usingconventional radiation doses are poor, and the local recur-rence rate is high. Tumor control can probably be improvedby increasing the dose (e.g., 1, 2). However, the surrounding

healthy lung tissue and the esophagus are dose limiting(e.g., 3, 4). To enable dose escalation, the irradiated sur-rounding normal tissue volume should be minimized. Onepossible approach is to reconsider the margins for the con-ventional planning target volume (PTV). The PTV is gen-erally defined as the clinical target volume (CTV) (visibletumor plus margin for microscopic extensions) plus a mar-

gin to account for geometric uncertainties (5). The mostimportant interfractional geometric uncertainties are patientsetup errors, tumor shrinkage or growth, and respiratorybaseline shifts (i.e., shifts in respiration levels). Intrafrac-tional geometric uncertainties are due to respiratory andcardiac motion. Eliminating geometric uncertainties allows

for a reduction of the CTV to PTV margins, reducing thevolume of irradiated normal tissue and normal tissue com-plication probability and enabling dose escalation.

A single free-breathing computed tomography (CT) scanis often used for radiotherapy planning for lung tumors.However, respiration-induced tumor motion (TM) duringacquisition causes artifacts in tumor shape and position (6,

Reprint requests to: Eugne M. F. Damen, Ph.D., Department ofRadiation Oncology, The Netherlands Cancer Institute, Antoni vanLeeuwenhoek Hospital, Plesmanlaan 121, Amsterdam 1066 CX, TheNetherlands. Tel: (31) 20-512-2205; Fax: (31) 206691101;

E-mail: [email protected] by Grant NKI 03-2943 from the Dutch Cancer Society.

AcknowledgmentsThe authors thank Jasper Nijkamp for his clin-ical software development and Harry Bartelink, Ben Mijnheer, andLeah McDermott for critical reading of the manuscript.

Received Jan 6, 2006, and in revised form April 10, 2006.

Accepted for publication April 10, 2006.

Int. J. Radiation Oncology Biol. Phys., Vol. 65, No. 5, pp. 15601571, 2006Copyright 2006 Elsevier Inc.

Printed in the USA. All rights reserved0360-3016/06/$see front matter

1560


2/12

7). The cause of these artifacts is that the CT scanneracquires a stack of images without time information fromthe moving tumor, thus obtaining a set of arbitrary snap-shots of moving structures. To overcome this problem,time-resolved four-dimensional (4D) scanning techniqueswere developed (813). Basically, all these methods aresimilar i.e., the acquired oversampled three-dimensional

(3D) data are sorted by the patients respiratory phase usingan external breathing signal, yielding a set of 3D reconstruc-tions at different breathing phases. This set (i.e., 4D imagedata) provides temporal and spatial motion information thatcan be used to optimize treatment planning.

Currently, however, available commercial treatment plan-ning systems cannot handle a 4D-CT data set as input fortreatment planning. Several strategies for the implementa-tion of 4D data sets in the treatment planning process havebeen published:

Slow CT scan: Srensen de Koste et al. (14) used a slow CTscan (which can be considered as the time average of the

4D data set) to delineate the target volume encompassingthe tumor at any position during the breathing cycle(internal target volume). A 5 mm expansion was added tothe internal target volume to cover the CTV and a 5-mmexpansion was added from the CTV to the PTV.

Maximal inhale and exhale composite delineation: Allen etal. (15) studied the delineation of the gross tumor volume(GTV) in a maximal inhale scan and a maximal exhalescan and created a composite of the two delineation setsfor planning (GTV to CTV margin recipe was not given).

All-phase composite segmented tumor: Rietzel et al. (16)and Underberg et al. (17) proposed a method (based onmaximal intensity projection) that automatically creates acomposite segmented CT scan of all respiratory phasescans (the full 4D set). From this composite CT scan, thecomposite GTV was delineated (GTV to CTV marginrecipe was not given).

Breath-hold CT scan (with active breathing control): Wonget al. (18) and Rosenzweig et al. (19) studied breath-holdCT acquisition using the active breathing control deviceand voluntary breath-hold, respectively, to obtain a single3D scan without motion artifacts for treatment planningpurposes (GTV to CTV margin recipe was not given).

Deformation dose mapping: Keall et al. (20) proposed a 4Ddose calculation method based on 4D deformation mapsfrom the 4D-CT scan. The deformation maps were ap-plied to the peak inhale delineations of the tumor andother structures. For each respiratory phase of the 4D set,an optimized treatment plan was calculated.

The first three methods can be summarized as compositetarget volume concepts, in which the target volumes areoften enlarged. Moreover, delineation of moving structuresin the slow CT method is difficult because of motion blurring.The second and third method need multiple reconstructions fortumor delineation and dose calculation. The breath-hold

method is not applicable to all lung cancer patients because oftheir physical condition. Moreover, breathing control also

needs to be maintained during treatment. The effectiveness ofthe last method, by Keall et al. (20), was demonstrated in afeasibility study, but its clinical implementation is not yetpossible because of hardware limitations.

In this report, we have proposed the selection of a singlewell-chosen CT scan from a 4D set. The choice of the singleCT scan was based on studies by Engelsman et al. (21) and

Witte et al. (22). These studies showed that if the tumor isirradiated at its average position during the respiration cy-cle, because of the presence of the wide-beam penumbra inthe lung, good dose coverage would still be obtained even ifthe tumor was not fully within the high-dose region for asmall part of the breathing cycle (0.6% and 6% tumorcontrol probability loss for 5 mm and 15 mm motion am-plitude with 0 margin, respectively, and no setup errors).Such an approach makes margin reduction possible.

The aim of this study was to eliminate the systematicerrors in the imaging process induced by respiratory motionand to obtain a more representative scan for delineation of

the target area, normal structures, and dose calculation. Inthe presented approach, motion was not taken into accountwhen delineating the tumor (thus no internal target volume)but will be incorporated in the margin expansion from CTVto PTV. This margin included the dose blurring caused byrespiratory motion during treatment. See Fig. 1 for theclinical protocol for 4D analysis and scan selection cur-rently used in our hospital

METHODS AND MATERIALS

Patient group

Patients with a lung tumor who were scheduled for radiotherapyunderwent fluoroscopy to estimate the TM due to respiration. If thetumor was estimated to move 0.5 cm owing to respiration, thepatients were eligible for this study. All patients (12 men and 3women) included in the study gave written informed consent toparticipate. The local medical ethics committee approved the study.

Respiration-correlated 4D-CT imaging4D-CT scanning. During 4D-CT scanning, the patient was in-

structed to breath freely and normally. Patient respiration was regis-tered using a thermocouple (Type T, copper-constantan, S-CC-U-O-7/1, Volenec, Hradec Krlov, Czech Republic) inserted into the entryof a regular oxygen mask, which measures temperature changes in the

airflow during inhalation (cold) and exhalation (warm).During CT scanning, patients were positioned supine with their

arms raised above their head using an in-house developed forearmsupport. To obtain a position similar to that on the treatment couch,a flat tabletop (Sinmed, Reeuwijk, The Netherlands) was placed onthe couch of the CT scanner (20-slice Somatom Sensation Open,Siemens, Forchheim, Germany). The helical cardiac scanningmode of the CT scanner was used for the respiration-correlatedimaging. However, the thermocouple respiratory signal was usedfor data sorting, instead of the cardiac input signal.

The scans were made at 120 kV, 1000 eff.mAs. The completethorax was scanned (3035 cm) from 5 cm above the top of thelung down to 5 cm below the diaphragm in the inhale position. The

scan time was 6070 s.Scan reconstruction. The beam-on signal of the CT scanner was

1561Clinical implementation of 4D-CT scans J. W. H. WOLTHAUS et al.


3/12

used to synchronize the respiration signal with the CT data. Thetime delay (response time) between the movement of the internalstructures of the lung and the external (thermocouple) breathingsignal (0.4 s, see Appendix) was corrected by back shifting therespiratory signal with respect to the beam-on signal. After thistime-delay correction, the respiratory signal should, in principle,be in phase with the CT data of the internal structures. The

respiration-correlated protocol on the scanner, an adaptation of thecardiac protocol, uses the peaks (maximal inhalation) of the res-piration signal to sort the raw CT data. Between the peaks, therespiratory cycle is divided into 10 equidistant time-percentagebins (0% at maximal inhalation to 90%). Within one cycle, thetime bins are equidistant, but from cycle to cycle, the exact lengthof a certain bin may vary owing to variations in the breathingperiod. Using equidistant time bins, which is equal to linear datasampling, the asymmetry between inhalation and exhalation lengthis incorporated in the fourth dimension of the resulting 4D scan byhaving a different number ofgenerally more3D-CT frames inthe exhalation phase than in the inhalation phase.

For each table position (every 3 mm) and time-percentage

within the breathing cycle, the corresponding location in the CTsinogram (as a function of table position and gantry angle) was

determined. A slice was reconstructed using data from 110 to110 (180 plus fan angle) from the determined gantry angle.The number of slices within one time bin was approximately 100for the chosen slice thickness of 3 mm.

Choice of CT acquisition parameters. The quality of 4D scan-ning in the helical mode depends critically on the interplay be-tween the breathing characteristics of the patient and the acquisi-

tion parameters used for scanning. According to Ford et al. (9),complete coverage of moving structures over a full breathingcycle is obtained when CLGantry-rotation pitch CLRespiration,where CLRespiration is the breathing cycle length of the patient andCLGantry-rotation is the gantry rotation time. If the CLGantry-rotation isshorter than the pitch CLRespiration, undersampling of the datawill occur, leading to gaps in the 4D data for particular phases. Ifthe CLGantry-rotation is longer, oversampling will occur and motionwill be blurred in the slice. The Somatom 20-slice scanner allowsgantry rotation times of 0.5 and 1.0 s. We therefore used twodifferent sets of acquisition parameters. When the averageCLRespiration 1 standard deviation was less than 5 s, aCLGantry-rotation of 0.5 s and pitch of 0.1 was used, otherwise a

CLGantry-rotation of 1.0 s and pitch of 0.15 was used. This procedureprevented undersampling of a breathing period up to 6.7 s. The

at the CT scanner

oxygen mask

with thermocouple

is placed on the patient

pre-acquisition of

respiration

signal

compute

mean & std.dev.

cycle length

std.dev.

cycle length?

mean+std.dev.

cycle length?

set CT gantry rotation

to 0.5 s

pitch = 0.10

set CT gantry rotation

to 1.0 s

pitch = 0.15

scan patient

reconstruct scan

> ~ 2 s

< ~ 2 s

> 5 s< 5 s

after scanning

examine tumor motion

(TM) of 4D scan

visually

cranio-caudal

TM amplitude?

determine

diaphragm motion

determine

(cranio-caudal)

tumor motion

compute mid-ventilation

time-percentage

during exhale

reconstruct

MidV CT scan

> ~ 2 cm< ~ 2 cm

verify visually in

green/purple composite

viewer if scan is adequate

Fig. 1. Considerations presented in this report can be summarized in the following clinical protocol currently in use at ourinstitution. The protocol is divided into two parts: during and after scanning. 4D four-dimensional; MidV mid-ventilation.

1562 I. J. Radiation Oncology Biology Physics Volume 65, Number 5, 2006


4/12

degree of oversampling may vary in a single 4D scan, depending

on the variability of the breathing cycle length of a patient duringscan acquisition.

Concept and definition of mid-ventilation scansTo plan the radiation of the tumor in the mean tumor position, we

reconstructed a single 3D CT scan from the 4D data set that representsthe mobile structures close to their (time-weighted) mean position.This scan was called the mid-ventilation (MidV) CT scan and wasused for delineation of the tumor and normal tissue, as well as fortreatment planning dose calculation. First, we defined the mean po-sition and mid-ventilation time-percentage (t%midv) (Fig. 2).

Definition. The mean position (TM) of a moving tumor or struc-

ture is the time-weighted mean position of the center of gravity of

that tumor or structure in all three dimensions (leftright [LR],cranio-caudal [CC], and anterior-posterior [AP]). If hysteresisoccurs in the TM (23), this mean position is not necessarily on thetrajectory of the object (Fig. 2a).

Definition. Mid-ventilation time-percentage . . . the t%midv isthat time/percentage in the respiration cycle at where the tumor is

closest to the time-weighted mean position. In general, two solu-tions are possible: the mid-ventilation time-percentage can bedefined during exhalation and inhalation (Fig. 2a).

When the tumor moves in more than one direction (hysteresis),it might be possible that there is more than one closest distance

between the tumor and the mean position TM (Fig. 2a). Toovercome this ambiguity in the determination of the t%midv, only

the CC motion curve was considered (Fig. 2b,c). The CC respira-tory movements were usually dominant compared to the LR and

AP movements. The CC motion curve does not give the real

minimal distance between the tumor and mean position, but pro-vides an unambiguous single t%midv for exhalation and inhalation.As a consequence of this approach, two types of errors are intro-duced: first, a geometric error with respect to the mean position(h), due to hysteresis in the TM; and second, an error in thedetermination of the time-percentage (midvt%), owing to simplifi-cation from using the CC motion direction only. Because of hyster-esis, phase shifts exist between the motion curves of the differentdirections (Fig. 2c, lower panel), resulting in a geometric error h forthe other directions.

To calculate the t%midv, the TM was calculated using the nor-malized (scaled between 0 and 1) CC motion curve (as inferredfrom Determination of TM by tumor tracking [see below]),

indicated by the horizontal solid line in Fig. 2. Subsequently, themotion curve was interpolated using cubic spline interpolation.The t%midv in the exhalation phase (t%midv-exhale) and inhalationphase (t%midv-inhale) were determined. Finally, the correspondingMidV-CT scan was reconstructed (at t%midv-exhale or t%midv-inhale).

Verification of MidV-CT scan selection. To check whether thisreconstructed MidV-CT scan was acceptable, the scan was verifiedby visually inspecting the position of the structures of theMidV-CT scan with respect to the corresponding moving struc-tures in the 4D-CT scan using an in-house developed compositeviewer (Fig. 3). In this composite viewer (24), the MidV-CT isrepresented by the color purple (magenta) and the 4D-CT scan isrepresented by the color green. When overlapping, regions of

corresponding density are then represented in gray scale; un-matched regions will preserve their color. Verifying the geometric

Tumor trajectory

t%midv-inhale

0%

TM

(b)

t%midv-exhale

Simplification (c) Simplification

100%0%

Maximum

inspiration

Maximum

expiration

mean

position

t%midv-inhalet%midv-exhale

Cra

nio-caudalprojection

(a) Concept

Tumor trajectory

t%midv-inhale

0%

TM

t%midv-exhale

Maximum inspiration

Maximum expiration

100%0%

Maximum

inspiration

Maximum

expiration

mean

position

t%midv-inhalet%midv-exhale

Anterior-posterior

projection

(TM)

(TM)

Fig. 2. (a) Schematic overview of trajectory of tumor with its mean position TM and multiple mid-ventilation positionsin exhalation (t%midv-exhale) and inhalation (t%midv-inhale) due to hysteresis (multiple closest distances between tumor andmean tumor position). Geometric error with respect to mean position denoted by h. (b) Simplification of concept byconsidering craniocaudal (CC) movements only. Error in determination of time-percentage midvt% , due to simplificationof only using CC motion direction; (c) CC projection of tumor trajectory, which is used to estimate mid-ventilationtime-percentages. Because of hysteresis, phase shift exists between motion curves of different directions (c, lowerpanel), resulting in geometric error h for other directions.



5/12

average with the human eye is a good first estimate of the time-weighted mean position.

Determination of motionDetermination of TM by tumor tracking. The TM in the 4D-CT

scan was determined using an image registration procedure (12). Avolume-of-interest (VOI) was defined around the tumor in a ref-erence CT frame (0%) using a manually drawn mask, encompass-ing the tumor. The VOI was subsequently registered to the scansof the other time-percentages on the basis of the gray value usingthe correlation ratio (25) of all voxels in the VOI. This procedurewas repeated three times for three different reference CT frames(0%, 30%, and 70%). From each TM curve, the mean tumorposition was subtracted to obtain the relative motion curves. Thethree curves were averaged to reduce spurious results due torandom drawing or matching errors, resulting in one (relative)motion curve that was used for further analysis (Fig. 4). The TMcurves were considered to be the reference.

Automated determination of diaphragm motion. Automated de-termination of the tumor trajectory can be complex, especially ifsegmentation of the tumor is not available. Moreover, if the tumoris not solitary but is attached to the mediastinum or thoracic wall,tumor segmentation becomes difficult. We therefore investigatedwhether the diaphragm motion (DM) derived from the imageinformation in the 4D-CT scan could be used as a surrogate for CCTM. This method is fully automated after the tumor has beenlocated in the right or left lung. First, the first frame (0% timepercentage) from the 4D scan is selected. This scan is projectedon an axial plane (Fig. 5a) and subsequently projected on the LRaxis obtaining a one-dimensional profile of the pixel intensities(Fig. 5b). The position of the maximum is used to distinguish

the left from the right lung (Fig. 5c). Because the patient in theexample ofFig. 5 had a tumor in the left lung, we considered thelower left part of the 4D-CT scan (Fig. 5d). This selection wascropped to the lung region, subsequently cropped to the lower halfof the lung, and taken as the VOI. An average CT VOI wasobtained by averaging the 10 time-percentages of the 4D-VOI. Thefirst (single) CT frame VOI (0%) was then subtracted from theaverage VOI (Fig. 5e,f). Finally, all voxels of the subtracted VOIs

were averaged over space (Fig. 5g,h). The steps e through gwere repeated for all 10 CT frames, and the 10 averaged valuesresulted in the motion curve of the diaphragm (Fig. 5h). The DMcurve was scaled between 1 and 0 for comparison with the TM.The t%midv and the corresponding mid-ventilation tumor positionderived from the DM were compared with the mean tumor positionas inferred from analysis of the TM (resulting in error DM).

Validation of external respiration signalPrevious studies (26, 27) have reported that breathing measure-

ments using an external device correlate well with movements ofthe internal structures, but phase shifts and time delays may existbetween the two signals. Therefore, this phase shift and time delaywas determined for the equipment used in our study. The verifi-cation of the thermocouple system, postprocessing of the thermo-couple signal, and correlation to the DM is described in theAppendix. The postprocessed thermocouple signal, which is inphase with the DM, was used during 4D-CT reconstructions.

RESULTS

Respiration cycle length and amplitude

Fifteen patients were included in this study (six upperlobe, one middle lobe, and eight lower lobe tumors). Theaverage breathing cycle length was 4.4 s for all patients andover all cycles during scanning (Table 1). However, thecycle length varied between 1.8 s and 10.1 s. Seven patientshad a mean cycle length 1 standard deviation larger than5 s, and 10 of 15 patients had at least one cycle that waslonger than 5 s. Ten patients underwent scanning with the0.5 s gantry rotation protocol and five with the 1.0 s gantryrotation protocol. Three patients were breathing with ashorter cycle before the scanning than during the scanning;therefore, the choice of the scanning parameters was notoptimal. For 2 patients, a too-short gantry-rotation time wasaccidentally chosen (0.5 instead of 1.0 s). Although this

Fig. 3. Composite viewer method for visual verification of selectedmid-ventilation (MidV) computed tomography (CT) scan.MidV-CT is represented by purple (magenta) and four-dimen-sional (4D)-CT scan by green. When mixing the two scans, struc-tures that match become white and structures that do not matchwill preserve their color.

0 25 50 75 100

Translations(mm)

Time-percentage (%)

-10

-5

0

5

10

Maximum inspiration

(Left; Caudal; Anterior)

Maximum expiration

(Right; Cranio; Posterior)

Fig. 4. Example of tumor motion determined by automated gray-value matching algorithms. Solid, dashed, and dottedline denotes cranio-caudal, leftright, and anterior-posterior movements, respectively.



6/12

affected the scan quality, it did not seriously hamper thequantitative measurements (e.g., artifacts were not presentin all 4D-CT frames and the resulting motion curves weresmooth). CC amplitudes (Table 1) of the TM, as inferredfrom analysis of TM (see the section Determination of TMby Tumor Tracking above) ranged from 2 to 26 mm (mean

12.3 7.1 mm).

Thermocouple time delay

The mean thermocouple time delay (see Appendix), mea-sured over approximately 6 80 breathing cycles, was0.40 0.18 s. This delay was equipment dependent andmust be determined for the equipment used. In addition, itappeared that a very weak correlation was present between

the cycle length and time delay (correlation coefficient 0.22).

Mid-ventilation

Figure 6 shows a typical example of the DM curve withthe corresponding CC TM curve. The DM correlated highlywith the TM.

The average t%midv according to the CC TM (referencestandard) was 17.4% 5.5% for exhalation and 75.0% 6.5% for inhalation (Table 2). The difference in t%midvbetween exhalation and inhalation did not appear to corre-spond to the scheme presented in Fig. 2c, which was be-

cause of an unexpected phase shift (see the section Mid-Ventilation Determination under Discussion). To make

Fig. 5. Schematic overview of determination of diaphragm motion. First frame (0% time-percentage) from four-dimensional (4D) scan was picked. This scan was projected to axial plane (a) and subsequently projected to leftright

axis obtaining one-dimensional profile of pixel intensities (b). Position of maximum (max.) used to distinguish left fromright lung (c). Because patient shown had tumor in left lung, we considered left part of 4D-CT scan (d). This selectionwas cropped to lower half of lung region. 4D cropped image was averaged in time (e). Then, a single CT frame wassubtracted from the average image (f). Finally, all voxels in subtracted images were averaged over space (g). The stepse through g were repeated for all 10 CT frames, resulting in (h) motion curve.

Table 1. Patient breathing characteristics (n 15)

Mean cycle length (s)

TM Amplitude (mm)

LR CC AP

Mean 4.4 0.7 2.2 12.3 3.6SD 1.2 1.2 7.1 2.2

Abbreviations: AP anterior-posterior; LR leftright; CCcranio-caudal; TM tumor motion.

See 4D-CT Scanning and Determination of TM by TumorTracking under Methods and Materials for details.



7/12

the geometric results of the MidV-CT concept comparablewith other geometric error contributions in radiotherapy, hand DMwere calculated as the mean (group mean) and thestandard deviation () over the patient group. The geomet-ric error, h, with respect to the mean position (Table 2)was, for exhalation, 0.2 0.3 mm in the LR direction and0.7 0.8 mm in the AP direction. For inhalation, it was0.0 0.4 mm in the LR direction and 1.1 1.1 mm in theAP direction. By definition, h in the CC direction was 0

because of the one-dimensional CC determination oft%midv.Using the DM method, the average t%midv for exhalationwas 19.9% 4.1% and for inhalation was 76.7% 5.3%(Table 3). The difference between the DM and TM methodswas significant (exhalation and inhalation p 0.01, pair-wise t test), but very small relative to the bin size of 10%(exhalation 2.5% 3.5%). The maximal geometric error,DM, with respect to the mean position (Table 3) in the CCdirection, was0.2 0.8 mm for exhalation and 0.6 1.5mm for inhalation. The corresponding values in the AP andLR directions for exhalation and inhalation were smallerthan 1.1 1.1 mm, not significantly different from the

reference of TM (p 0.1). Correcting for the systematic

time-percentage difference between DM and TM did notresult in a smaller geometric error with respect to the meantumor position (DM).

DISCUSSION

The results of this work have shown that the mid-venti-lation concept is a useful solution to incorporate motioninformation from 4D-CT scans in the treatment planningprocess. The determination of the t%midv using TM and DMis accurate; therefore, the appropriate planning CT scan canbe selected from the 4D data set. The use of a single CTscan (with the tumor in the mean position) allows formargin reduction, resulting in a smaller irradiated volume(see the section Potential of Reduced Margins Because ofImproved Imaging).

Thermocouple time delay

The comparison of the thermocouple signal vs. DMshowed that the delay time of the thermocouple is stable andequipment dependent (see Appendix). If this time delay isnot corrected, a time-percentage difference t% will existbetween the first frame of the 4D-CT scan (0%) and theactual maximal inhalation. t% can be estimated (afterreconstruction of the 4D-CT scan) using DM determination,as described in our report. However, correction for this timedelay before CT reconstruction is important, because thewrong synchronization of the thermocouple signal to the

raw CT data will result in incorrect sorting of data intorespiratory phases. This can be clarified in a worst-casescenario example: when a patient would be scanned duringtwo subsequent breathing cycles, with a cycle length of 3and 6 s, respectively, and if a thermocouple time delay of 3 sexists that has not been corrected for, the first half of the CTdata of the second breathing cycle would be sorted over afull cycle. This type of error in sorting can result in seriousimaging artifacts (e.g., when using external thorax move-ment respiration signal, see the section Phase Shifts due toImperfections of 4D-CT below). However, the delay timeof our thermocouple equipment (0.4 s) and the standard

deviation of the cycle length were small (Table 1) comparedwith the breathing cycle length, resulting in negligibly small

Fig. 6. Example of high correlation between cranio-caudal tumormotion (solid curve) and diaphragm motion computed by imagegradient of four-dimensional scan (dashed curve). Both motion curveswere scaled between maximal inhalation (1) and maximal exhalation(0). Note, time scale runs from 0% to 200% to emphasize periodicity.

Table 2. Average mid-ventilation time-percentages (t%midv)based on craniocaudal tumor motion and corresponding errors in

three directions due to hysteresis (h)

t%midv (%)

h (mm)

Exhale Inhale

Exhale Inhale LR CC AP LR CC AP

Mean 17.4 75.0 0.2 0.0 0.7 0.0 0.0 1.1SD 5.5 6.5 0.3 0.0 0.8 0.4 0.0 1.1

Abbreviations as in Table 1.

Table 3. Average mid-ventilation time-percentages (t%midv)based on automated diaphragm motion (DM) method

t%midv (%)

DM (mm)

Exhale Inhale

Exhale Inhale LR CC AP LR CC AP

Mean 19.9 76.7 0.1 0.2 0.8 0.0 0.6 1.1SD 4.1 5.3 0.4 0.8 0.9 0.4 1.5 1.1

Abbreviations as in Table 1.Geometric error is denoted by TM between determined mid-

ventilation position and real mean position of tumor during breath-ing for diaphragm motion method.



8/12

artifacts in the 4D image reconstruction, even if the delaywas not corrected.

Mid-ventilation determination

Phase shifts due to imperfections in 4D-CT. Because the4D-CT data were sorted between the two maximal inhala-

tion peaks after correcting for the thermocouple time delayas described in the section Scan Reconstruction, the firstCT frame (0%) of the 4D set was expected to represent themaximal inhalation phase. However, in general, a smalltime-percentage difference (t%) was still found betweenthe first frame of the 4D-CT scan (0%) and the actualmaximal inhalation phase. The time-percentage difference(t%) was 3.3% 5.1% using TM and 1.0% 4.1% usingthe DM method.

A possible cause for t% could be the quantization orbinning error, which was 5% for the low number of bins(10) used during reconstruction. A second possible cause

could be the existence of a (small) phase variation betweeninternal structures and the external respiration system (seesection Thermocouple Time Delay). Finally, the use of aone-dimensional (CC) simplification could be a third causeof the determined phase shift, because it does not have torepresent the maximal inhale position in three dimensions.

Rietzel et al. (16) reported a significant phase differencebetween external thorax movements (from the Varian RPMrespiratory measurement system) and internal structures. Ifthis phase difference is stable during the entire 4D scan acqui-sition, the phase difference can be found retrospectively bymethods we have described. However, when patients switch

from lower body (abdomen) to upper body (chest) respira-tion during scanning, a significant phase change may occur,which can result in severe artifacts on the 4D images.

Mid-ventilation time-percentage. The t%midv derivedfrom the TM and DM curves appeared to be significantlydifferent (Tables 2 and 3). However, relative to the bin size(10%), the difference was small; therefore, the automatedDM method can be used as a surrogate for the MidVestimation with the TM method. Moreover, after correctingthe t%midv for patients individual time-percentage differ-ence t% (thus, setting the maximal inhalation to 0%, seeprevious section), the t%midv was no longer very different.

The t%midv was 20.7% 2.6% and 20.7% 2.2% and78.7% 2.9% and 78.8% 2.9% in exhalation and inha-

lation for TM and the DM method, respectively. The stan-dard deviation in t%midv was reduced, supporting the use ofthe DM method as a surrogate for TM. These results of thet%midv were in agreement with the asymmetric properties ofrespiration (t%midv not 25% or 75%), as described by Sep-penwoolde et al. (23). It appeared that the TM and DMcurves were in phase except for patients with tumor attached

to the mediastinum or chest wall. In those cases, the curveswere also slightly different in shape, resulting in a differentt%midv for TM and DM. TM will be used to select theMidV-CT scan.

Uncertainties of mid-ventilation with respect to mean

tumor position. From the standard deviation in h and DM(Tables 2 and 3), it can be concluded that a MidV recon-struction in exhalation is more accurate than that withinhalation. Because the exhalation phase is longer than theinhalation phase, the excursion per time-percentage isgreater during inhalation than during exhalation. A smalluncertainty in t%midv will thus result in a larger geometric

error at inhalation than at exhalation.For clinical practice, a cutoff value in the TM amplitude,below which the automated DM method can be safely used,might be beneficial. First, for all patients, the mean andstandard deviation of the geometric error relative to theamplitude was computed in the three directions for allpatients. This relative error (rel) was 0.14 0.04 mm/mm,0.07 0.04 mm/mm, and 0.31 0.10 mm/mm for the LR,CC, and AP directions, respectively. Limiting the meangeometrical error 1 standard deviation to 2 mm for alldirections can be described as a constraint:

max rel,LRALR; rel,CCACC; rel,APAAP 2 mm

Using the statistics of the amplitude for this patient group(Table 1), the proportions of the mean amplitudes in the LRand AP direction relative to the CC amplitude were

ALR 0.18ACC and AAP 0.29ACC

Applying these amplitude relations in the constraint equa-tion showed that the use of the automated DM method islimited to a maximal CC tumor excursion of 2 cm.

Comparison of MidV-CT with conventional CT. To eval-uate the methods for the construction of a treatment plan-ning CT scan, we compared the results with those fromconventional CT scanning (no respiration-correlated imag-ing). Because we had determined the TM for all directionsand all patients, we could easily estimate the errors inconventional imaging (conv), which is the standard devia-tion of the motion over the respiratory cycle (28). The meanis 0 by definition. The standard deviation for all patients was0.8 mm for LR, 3.9 mm for CC, and 1.5 mm for AP (Table4). This is in agreement with the conclusions of van Herk(28), that the geometric error with respect to the mean tumor

position (conv) was one-third of the peak-to-peak ampli-tude. Comparing the DM method to conventional scanning,

Table 4. Mean error, calculated for patient group of presentstudy, when scanning mobile tumors with conventional

free-breathing 3D method

conv (mm)

LR CC AP

Mean 0 0 0

SD 0.8 3.9 1.5

Abbreviations: 3D three-dimensional; other abbreviations asin Table 1.

Error calculated in three directions; group mean was 0 by definition.



9/12

the geometric variation DM was four times smaller in theCC direction than conv (Table 3).

Dose verification of the new MidV concept is beyond thescope of this report; however, we refer to Cho et al. (29)who showed that, with constant error margins, small geo-metric variations have a negligible influence on the dosedistribution. The application of the results of Cho et al. (29)

to the relatively small geometric differences between con-ventional and MidV-CT scans showed that MidV-CT scanscan safely replace conventional CT scans. The improvementgained by the use of MidV-CT will primarily be a reductionin the error margins. A clinical protocol based on thepresented results concerning 4D scanning and mid-ventila-tion CT reconstruction is given in Fig. 1.

Reproducibility of respiratory patterns. Reproducibilityin respiratory patterns during 4D imaging and treatmentdelivery is an important issue. We have begun a study of thepossible changes in breathing patterns and mean tumorposition when using an oxygen mask. Additional 3D scans

are made before treatment with a cone-beam systemmounted on a linear accelerator (30). These scans can bereconstructed into 4D scans (31) and mean tumor positionand amplitude determined. Although the preliminary resultshave shown no significant differences in breathing patternswith or without a mask, the final conclusions cannot yet bemade.

Influence of breathing cycle length on

4D-CT reconstructions

The small standard deviation in the average cycle length(Table 1) suggests that the patients were breathing regularly

during scanning, resulting in stable and practically artifact-free 4D-CT reconstructions. This makes a choice for one ofthe two protocols (0.5 s or 1.0 s gantry rotation) justified.However, a lower pitch or more choices of gantry rotationspeed would be desirable to overcome the remaining arti-facts. Interpolation artifacts in the 4D image appeared atlocations where undersampling occurred. If the number ofthese artifacts were small, tumor match and DM measure-ments would not be seriously hampered. In some cases, thepatient was breathing faster during setup than during theactual 4D scanning. Consequently, the gantry rotation wastoo fast compared to the patients cycle length, causing

multiple imaging artifacts and making delineation of struc-tures or tumor more difficult. However, quantitative mea-surements were not noticeably influenced. For patients withirregular breathing, one solution was to wait a couple ofminutes until the respiration stabilized, before measuringthe cycle length, which was included in the 4D scanningprotocol. In addition, audiovisual breathing coaching canstabilize the breathing signal amplitude and period (32),which will improve the image quality of the MidV-CTscans. However, coaching is often not applicable to thispatient group because of physical limitations. Moreover, theinfluence of coaching on the baseline (mean respiration

level) variation is still unknown. Amplitude-based CT databinning (33) during regular breathing would not result in

better image quality and has the drawback that time infor-mation is lost. Time information is necessary to calculatethe time-weighted mean position. Furthermore, with irreg-ular breathing, some amplitude bins can be sparsely filled,resulting in serious artifacts (gaps).

A principle disadvantage of our proposed method is thatthe tumor has the greatest speed at mid-ventilation and can

result in more residual imaging artifacts than at maximalinhalation or maximal exhalation. However, according toRietzel et al. (27), these artifacts are smaller than the arti-facts due to irregular breathing. Moreover, compared withconventional scanning, these speed-induced artifacts arenegligible.

Potential of reduced margins because of

improved imaging

Treatment margins (from GTV to PTV) can be calculatedusing the margin recipe of van Herk et al. (22, 34, 35):margin 2.5 0.7 , where and denote the standard

deviations of the systematic and random errors, respec-tively. These errors consist of contributions from respiration(periodic motion and mean tumor position) and setup errors(errors due to cardiac motion were not considered). Delin-eation uncertainties also contribute to additional margins(36) and are expected to become smaller using 4D treatmentplanning because of improved visualization of the tumorshape. However, we did not consider detailed knowledge ondelineation uncertainty using free-breathing scans (37, 38).

Margins due to systematic and random errors were esti-mated for a patient with a spherical tumor of 20 mm and a20 mm CC tumor movement in diameter in two cases (Table

5): without using 4D imaging (columns 2 and 3) and withusing 4D imaging (column 4 and 5). The displaced tumor inthe conventional planning CT (random snapshot of thetumor) with respect to the mean position of the moving

Table 5. Estimated magnitude of existing uncertainties incraniocaudal direction for two cases: conventional 3D-CT or

4D-CT scan for treatment planning

Conventional3D-CT 4D-RCCT

(mm)

(mm) (mm)

(mm)Respiration (A

CC

20 mm)Periodic motion 6.7 6.7 0 6.7Baseline variation 2 2 2 2

Setup error 1.5 3.1 1.5 3.1Total 7.2 7.6 2.5 7.6Margin (CTV to PTV) 23.2 11.6

Abbreviations: 3D three-dimensional; 4D four-dimen-sional; A

CC amplitude of tumor motion in cranio-caudal direc-

tion; CTV clinical target volume; PTV planning target volume.Required margin from CTV to PTV was calculated according to

van Herk et al.: margin 2.5 0.7 , where and denote

standard deviations of distribution of systematic and random error,respectively.



10/12

tumor is referred to as the systematic contribution of theperiodic motion. The probability distribution of the tumor ina certain position during treatment, which has a standarddeviation of one-third of the peak-to-peak amplitude (28), isreferred to as the random contribution of the periodic mo-tion. The systematic and random setup errors were set at 1.5mm and 3.1 mm, respectively, in the conventional and 4D

MidV situation using portal images (39).Using 4D imaging will theoretically reduce the system-atic contribution due to breathing motion to 0. The randomerror distribution due to respiration will not change becauserespiration is still present during treatment. However, themargin for random errors might possibly be reduced, suchas was shown in simulations due to the shallow penumbra inlung tissue (21, 22). The additional margins needed tocompensate for the breathing motion are small if the irra-diated volume is centered at the average position of thetumor using 4D imaging techniques. Variations in the av-erage position of the tumor (baseline) give a contribution

that still remains in 4D treatment planning. This contribu-tion can only be reduced when using an off-line 4D verifi-cation protocol (using multiple 4D-CT scans), reducing thesystematic average position variation, or an on-line 4Dverification protocol (using multiple 4D-CT scans on thetreatment machine [40, 41]), reducing the systematic andrandom average position variation. Using multiple 4D cone-beam CT scans from 10 patients (42), this average positionerror was estimated to be 2 mm for both systematic andrandom errors. For the patient data in Table 5 with a CC TMof 20 mm, the use of 4D imaging would result in a reductionof the required margin in the CC direction with a factor of

2. With an amplitude in the LR direction of 4 mm and 6 mmin the AP direction (see the section Uncertainties of Mid-Ventilation with Respect to Mean Tumor Position), themargins in these directions will be 10 mm and 11 mm forconventional free-breathing CT scans and 9 mm and 9 mmfor MidV-CT scans, respectively. In this example (sphericaltumor of 20 mm diameter), this margin reduction decreases

the PTV volume from 14.4 cm3

to 8.3 cm3

(43% reduction).In addition, using the mean motion amplitude values ofTable 1, the average PTV reductions for tumors with diam-eters between 10 and 80 mm would be between 33% and12%. Finally, for the patient group used in this study, thecomposite target volume methods described in the Intro-duction (slow-CT and maximal intensity projection) wouldresult in an increased irradiated volume of 11% 3%(setup error and baseline variation as above). Our proposedMidV-CT method would result in a significant reduction of12% 11% irradiated volume.

CONCLUSION

We have developed a new concept for using 4D-CT scansto incorporate patient-specific motion information in radio-therapy for lung cancer. On the basis of the tumor motion,a single 3D-CT scan is chosen from the 4D-CT data set,which represents all lung structures in the time-averageposition. An alternative automated method was also devel-oped to select the single 3D-CT scan using the diaphragmmotion. Using mid-ventilation CT scans for treatment plan-ning instead of the conventional free-breathing CT scans,margin reduction is possible, which can reduce the treat-ment volume up to 50%.

REFERENCES

1. Belderbos JS, De Jaeger K, Heemsbergen WD, et al. Firstresults of a phase I/II dose escalation trial in non-small celllung cancer using three-dimensional conformal radiotherapy.

Radiother Oncol 2003;66:119126.2. Kong FM, Ten Haken R, Eisbruch A, et al. Non-small cell lung

cancer therapy-related pulmonary toxicity: An update on radia-tion pneumonitis and fibrosis. Semin Oncol 2005;32:S42S54.

3. Belderbos J, Heemsbergen W, Hoogeman M, et al. Acute esoph-ageal toxicity in non-small cell lung cancer patients after highdose conformal radiotherapy. Radiother Oncol 2005;75:157164.

4. Schwarz M, Alber M, Lebesque JV, et al. Dose heterogeneityin the target volume and intensity-modulated radiotherapy toescalate the dose in the treatment of nonsmall-cell lungcancer. Int J Radiat Oncol Biol Phys 2005;62:561570.

5. International Commission on Radiation Units and Measure-ments. Prescribing, recording, and reporting photon beamtherapy: ICRU report No. 50. Bethesda: ICRU; 1993.

6. Balter JM, Ten Haken RK, Lawrence TS, et al. Uncertainties inCT-based radiation therapy treatment planning associated with pa-tient breathing. Int J Radiat Oncol Biol Phys 1996;36:167174.

7. Chen GT, Kung JH, and Beaudette KP. Artifacts in computedtomography scanning of moving objects. Semin Radiat Oncol2004;14:1926.

8. Vedam SS, Keall PJ, Kini VR, et al. Acquiring a four-

dimensional computed tomography dataset using an externalrespiratory signal. Phys Med Biol 2003;48:4562.

9. Ford EC, Mageras GS, Yorke E, et al. Respiration-correlatedspiral CT: A method of measuring respiratory-induced ana-tomic motion for radiation treatment planning. Med Phys2003;30:8897.

10. Keall PJ, Starkschall G, Shukla H, et al. Acquiring 4D tho-racic CT scans using a multislice helical method. Phys Med

Biol 2004;49:20532067.11. Pan T. Comparison of helical and cine acquisitions for 4D-CT

imaging with multislice CT. Med Phys 2005;32:627634.12. Wolthaus JW, van Herk M, Muller SH, et al. Fusion of

respiration-correlated PET and CT scans: Correlated lungtumour motion in anatomical and functional scans. Phys MedBiol 2005;50:15691583.

13. Pan T, Lee TY, Rietzel E, et al. 4D-CT imaging of a volumeinfluenced by respiratory motion on multi-slice CT. Med Phys2004;31:333340.

14. Srensen de Koste J, Lagerwaard FJ, de Boer HCJ, et al. Aremultiple CT scans required for planning curative radiotherapyin lung tumors of the lower lobe? Int J Radiat Oncol Biol Phys2003;55:13941399.

15. Allen AM, Siracuse KM, Hayman JA, et al. Evaluation ofthe influence of breathing on the movement and modelingof lung tumors. Int J Radiat Oncol Biol Phys 2004;58:12511257.

16. Rietzel E, Chen GT, Choi NC, et al. Four-dimensional image-based treatment planning: Target volume segmentation and



11/12

dose calculation in the presence of respiratory motion. Int JRadiat Oncol Biol Phys 2005;61:15351550.

17. Underberg RW, Lagerwaard FJ, Slotman BJ, et al. Use ofmaximum intensity projections (MIP) for target volume gen-eration in 4DCT scans for lung cancer. Int J Radiat Oncol BiolPhys 2005;63:253260.

18. Wong JW, Sharpe MB, Jaffray DA, et al. The use of activebreathing control (ABC) to reduce margin for breathing mo-tion. Int J Radiat Oncol Biol Phys 1999;44:911919.

19. Rosenzweig KE, Hanley J, Mah D, et al. The deep inspirationbreath-hold technique in the treatment of inoperable nonsmall-cell lung cancer. Int J Radiat Oncol Biol Phys 2000;48:8187.

20. Keall PJ, Joshi S, Vedam SS, et al. Four-dimensional radio-therapy planning for DMLC-based respiratory motion track-ing. Med Phys 2005;32:942951.

21. Engelsman M, Damen EM, De Jaeger K, et al. The effect ofbreathing and set-up errors on the cumulative dose to a lungtumor. Radiother Oncol 2001;60:95105.

22. Witte M, van der Geer J, Schneider C, et al. The effects oftarget size and tissue density on the minimum margin requiredfor random errors. Med Phys 2004;31:30683079.

23. Seppenwoolde Y, Shirato H, Kitamura K, et al. Precise andreal-time measurement of 3D tumor motion in lung due to

breathing and heartbeat, measured during radiotherapy. Int JRadiat Oncol Biol Phys 2002;53:822834.24. van Herk M, De Jaeger K, de Munck J, et al. A delineation

system for N modalitiesSoftware aspects [Abstract]. Presentedat the ICCR 13th Heidelberg, Germany, 2000. p. 7375.

25. Roche A, Malandain G, Pennec X, et al. The correlation ratioas a new similarity measure for multimodal image registration.Proceedings of MICCAI98. Lecture Notes in Computer Sci-ence 1998; 1496:11151124.

26. Lu W, Parikh PJ, El Naqa IM, et al. Quantitation of the recon-struction quality of a four-dimensional computed tomographyprocess for lung cancer patients. Med Phys 2005;32:890901.

27. Rietzel E, Pan T, Chen GT. Four-dimensional computed to-mography: Image formation and clinical protocol. Med Phys

2005;32:874889.28. van Herk M. Errors and margins in radiotherapy. Semin RadiatOncol 2004;14:5264.

29. Cho BC, van Herk M, Mijnheer BJ, et al. The effect of set-upuncertainties, contour changes, and tissue inhomogeneities ontarget dose-volume histograms. Phys Med Biol 2002;29:23052318.

30. Jaffray DA, Siewerdsen JH, Wong JW, et al. Flat-panel cone-beam computed tomography for image-guided radiation ther-apy. Int J Radiat Oncol Biol Phys 2002;53:13371349.

31. Sonke JJ, Zijp L, Remeijer P, et al. Respiratory correlatedcone beam CT. Med Phys 2005;32:11761186.

32. Neicu T, Berbeco R, Wolfgang J, et al. Synchronized movingaperture radiation therapy (SMART): Improvement of breath-ing pattern reproducibility using respiratory coaching. Phys

Med Biol 2006;51:617636.33. Fitzpatrick MJ, Starkschall G, Antolak JA, et al. Displace-ment-based binning of time-dependent computed tomographyimage data sets. Med Phys 2006;33:235246.

34. van Herk M, Remeijer P, Lebesque JV. Inclusion of geometricuncertainties in treatment plan evaluation. Int J Radiat Oncol

Biol Phys 2002;52:14071422.35. van Herk M, Witte M, van der Geer J, et al. Biologic and

physical fractionation effects of random geometric errors. IntJ Radiat Oncol Biol Phys 2003;57:14601471.

36. van de Steene J, Linthout N, de Mey J, et al. Definition ofgross tumor volume in lung cancer: Inter-observer variability.

Radiother Oncol 2002;62:3749.37. Steenbakkers RJ, Duppen JC, Fitton I, et al. Reduction of

observer variation using matched CT-PET for lung cancerdelineation: A three-dimensional analysis. Int J Radiat Oncol

Biol Phys 2006;64:435448.38. Steenbakkers RJ, Duppen JC, Fitton I, et al. Observer varia-

tion in target volume delineation of lung cancer related toradiation oncologist-computer interaction: A Big Brotherevaluation. Radiother Oncol 2005;77:182190.

39. Erridge SC, Seppenwoolde Y, Muller SH, et al. Portal imag-ing to assess set-up errors, tumor motion and tumor shrinkageduring conformal radiotherapy of non-small cell lung cancer.

Radiother Oncol 2003;66:7585.40. Sonke JJ, Zijp L, Remeijer P, van Herk M, et al. Respiration-

correlated cone-beam CT. Med Phys 2005;32:11761186.41. Zijp L, Sonke JJ, van Herk M. Extraction of respiratory signal

from sequential thorax cone-beam X-ray images. Proceedingsof the XIVth ICCR, Seoul, Korea, 2004. p. 507509.

42. Sonke JJ, van Herk M, Belderbos J, et al. An off-line 4D conebeam CT based correction protocol for lung tumor motion[Abstract]. Int J Radiat Oncol Biol Phys 2005;63(Suppl. 1):S389S390.

APPENDIX

Thermocouple response time

To verify the functioning of the thermocouple system and to

determine the correlation with diaphragm motion, we mea-sured the diaphragm movements using projection images froma cone-beam CT scanner mounted on a linear accelerator (30)concurrent with the thermocouple respiration signal system for6 patients during 7 min. The respiration signal was acquiredusing an AD-converter simultaneously with the X-ray expo-sure pulses that were generated by the cone-beam imager. Theexposure peaks provided a time stamp for each projectionimage. These time stamps of the acquired projection imageswere synchronized to the thermocouple signal. Subsequently,the diaphragm height was determined automatically in eachprojection image (31, 41). These measurements do not require

a cone-beam CT system, but can be performed with conven-

tional fluoroscopy.Because the thermocouple (temperature) and diaphragm

position (distance) have different units, both signals werescaled between 0 (maximal exhalation) and 1 (maximalinhalation) as follows. An upper and lower envelope wascreated by drawing one line intersecting all peaks in thesignal and a second line intersecting all valleys, respec-tively. The original signals were scaled between their upperand lower envelopes. The advantage of this method is thatit is easy to calculate, preserves the asymmetric shape of thecurve (different length of exhalation and inhalation), andremoves a possible global trend. In Fig. 7, an example of the



12/12

two signals was plotted (dashed curve represents thermo-couple signal and solid curve represents diaphragm signal).

A time delay occurs in the thermocouple signal withrespect to the diaphragm (Fig. 7). The thermocouple timedelay is due to the warming up and cooling down of thethermocouple, the presence of stationary air (end of expi-ration, end of inspiration), and the response time of the

signal amplifier equipment. The mean thermocouple timedelay, measured for approximately 6 80 breathing cycles,was 0.40 0.18 s. This delay is equipment dependent andshould be determined for the equipment used. Consideringthe quantization error in the acquisition of a cone-beamprojection image (standard deviation of a uniform distribu-

tion acquisition time/ 12 0.371 12 0.107 s)and additional noise, the standard deviation is quite accept-able. From these data, it appeared that a very weak corre-lation was present between the cycle length and time delay(correlation coefficient 0.22), suggesting that little phasedelay exists and therefore the response time is mainly equip-ment dependent. This delay (determined with the cone-beam CT system) will be corrected during 4D-CT recon-structions on a standard CT scanner.

Fig. 7. Example of thermocouple respiration signal (dashed curve) anddiaphragm motion (solid curve) determined from cone-beam computedtomography projection images. Both signals were scaled between max-imal inhalation (1) and maximal exhalation (0). denotes time delaybetween two signals.


4dct rt planning red 2006

Documents