|Year : 2017 | Volume
| Issue : 6 | Page : 994-999
Investigating different computed tomography techniques for internal target volume definition
SA Yoganathan1, KJ Maria Das1, V Siva Subramanian2, D Gowtham Raj1, Arpita Agarwal1, Shaleen Kumar1
1 Department of Radiotherapy, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, Uttar Pradesh, India
2 School of Physics, Bharathidasan University, Tiruchirappalli, Tamil Nadu, India
|Date of Web Publication||13-Dec-2017|
Mr. S A Yoganathan
Department of Radiotherapy, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Rae Bareli Road, Lucknow - 226 014, Uttar Pradesh
Source of Support: None, Conflict of Interest: None
Purpose: The aim of this work was to evaluate the various computed tomography (CT) techniques such as fast CT, slow CT, breath-hold (BH) CT, full-fan cone beam CT (FF-CBCT), half-fan CBCT (HF-CBCT), and average CT for delineation of internal target volume (ITV). In addition, these ITVs were compared against four-dimensional CT (4DCT) ITVs.
Materials and Methods: Three-dimensional target motion was simulated using dynamic thorax phantom with target insert of diameter 3 cm for ten respiration data. CT images were acquired using a commercially available multislice CT scanner, and the CBCT images were acquired using On-Board-Imager. Average CT was generated by averaging 10 phases of 4DCT. ITVs were delineated for each CT by contouring the volume of the target ball; 4DCT ITVs were generated by merging all 10 phases target volumes. Incase of BH-CT, ITV was derived by boolean of CT phases 0%, 50%, and fast CT target volumes.
Results: ITVs determined by all CT and CBCT scans were significantly smaller (P < 0.05) than the 4DCT ITV, whereas there was no significant difference between average CT and 4DCT ITVs (P = 0.17). Fast CT had the maximum deviation (−46.1% ± 20.9%) followed by slow CT (−34.3% ± 11.0%) and FF-CBCT scans (−26.3% ± 8.7%). However, HF-CBCT scans (−12.9% ± 4.4%) and BH-CT scans (−11.1% ± 8.5%) resulted in almost similar deviation. On the contrary, average CT had the least deviation (−4.7% ± 9.8%).
Conclusions: When comparing with 4DCT, all the CT techniques underestimated ITV. In the absence of 4DCT, the HF-CBCT target volumes with appropriate margin may be a reasonable approach for defining the ITV.
Keywords: Computed tomography, cone beam computed tomography, four-dimensional computed tomography, internal target volume, lung cancer, respiratory motion
|How to cite this article:|
Yoganathan S A, Maria Das K J, Subramanian V S, Raj D G, Agarwal A, Kumar S. Investigating different computed tomography techniques for internal target volume definition. J Can Res Ther 2017;13:994-9
|How to cite this URL:|
Yoganathan S A, Maria Das K J, Subramanian V S, Raj D G, Agarwal A, Kumar S. Investigating different computed tomography techniques for internal target volume definition. J Can Res Ther [serial online] 2017 [cited 2020 May 28];13:994-9. Available from: http://www.cancerjournal.net/text.asp?2017/13/6/994/220353
| > Introduction|| |
The primary aim of the radiotherapy is to deliver maximum radiation dose to tumor and minimum radiation dose to surrounding normal tissues. However, achieving this aim is exigent due to various uncertainties associated with radiotherapy process. Respiratory motion is one such issue, which affects mainly thoracic and upper abdomen region. Respiratory motion facilitates the lung tumors to be dynamic and enables it to move in the order of 12 ± 2 mm in craniocaudal direction, 2 ± 1 mm in both anterior-posterior direction, and right-left direction. Hence, targeting the dynamic tumor with radiation beam poses additional challenges., The prognosis of inoperable lung cancer patients (Stage I/II non-small cell lung cancer) treated with protracted radiotherapy is poor with a 5-year survival rate of 13-39%, and the most common site of relapse is local failure. To improve the outcome, radiation dose escalation/altered dose fractionation scheme is investigated.,, One of the main limitations for these investigations is normal lung tissue toxicity, especially when using the radiotherapy with concurrent chemotherapy. Conformal dose sculpting is a promising method for sparing the normal tissues as much as possible without compromising dose to target. However, the real benefit of conformal radiotherapy may be negated by respiratory motion.
Recent developments in radiotherapy have resulted in various technical solutions for dealing the respiratory motion, i.e., motion encompassing,,,,,,, breath-hold (BH),, gating,,, and tracking ,,,, method. Detailed information about these methods can be found elsewhere. The motion encompassing method is the primary focus of this study and in this method, an internal margin is added to clinical target volume to create an internal target volume (ITV) for accommodating the physiological motion of target. There are two common approaches in motion encompassing method; population-based margin and patient-specific margin. Identical margin is given to a group of patients in the population-based margin (assuming “one size fits to all”), whereas customized margin as per tumor motion is given to individual patients in the patient-specific margin. Studies have shown that the patient-specific margins are advantageous over standard “population-based” margin.,
There are various techniques available for determining the patient-specific ITV, i.e. fluoroscopic method, slow computed tomography (CT),, BH-CT,, cone beam CT (CBCT),,, and four-dimensional CT (4DCT).,,,,
The purpose of this work was to experimentally evaluate the different CT and CBCT techniques for delineating the ITV. The accuracy of these ITVs was evaluated by comparing them with 4DCT ITVs.
| > Materials and Methods|| |
All the CT scans (fast CT, slow CT, and 4DCT) were acquired with helical mode (slice thickness 2 mm) using a wide bore multislice CT unit (Somatom sensation open, Siemens medical systems, Germany). Previously reported respiratory data of ten cases were selected, and the amplitude and breathing period of respiration are shown in [Table 1]. Three-dimensional (3D) sinusoidal and regular respiratory motion of target were simulated using a dynamic thorax phantom (CIRS, Norfolk, VA, USA). A lung equivalent rod containing a spherical target of diameter 3 cm was inserted into the lung equivalent lobe of the phantom [Figure 1]. This rod was connected to a motion actuator box which moves the rod for 3D target motion through linear translational and rotational motions. Because of the limitation of dynamic thorax phantom, the amplitudes of right-left and anterior-posterior motion which was >5 mm were reduced to 5 mm, and the amplitudes of superior-inferior motion which was >20 mm were reduced to 20 mm.
|Figure 1: Dynamic thorax phantom setup during cone beam computed tomography acquisition|
Click here to view
Fast CT scans were acquired using routine radiotherapy simulation protocol available on the scanner. Usually, to generate ITV, an external margin would be given around the target volumes of fast CT scan, but for straightforward comparison, no additional margins were added with fast CT target volumes.
Typical slow CT scan would be done with the gantry rotation speed equal to respiratory motion period. However, in modern helical CT scanners, it would not be possible to have the gantry speed 4 s/rotation (or equal to respiratory motion period). Recently, Chinneck et al. reported a novel method for acquiring slow CT scans using modern helical CT scanners which had the maximum rotation speed 1.5 s/rotation. By altering the helical scan gantry rotation time and pitch factor, the data acquisition time for each slice was effectively altered. They acquired the slow CT scan using the maximum gantry rotation speed of 1.5 s/rotation with the reduced pitch value of 0.5, and this enabled to have slices to be reconstructed over 3.0 s. Similar approach was adopted in our current investigation. In this study, the gantry rotation speed was kept at the maximum possible value, i.e., 1 s/rotation and the pitch was reduced to the minimum possible value 0.45. This enabled the data acquisition time for each image slice to be 2.22 s.
Recent advances in imaging resulted in state of art 4DCT for accurately defining the tumor trajectory. Numerous studies have compared the 4DCT target volumes with conventional CT volumes and showed that 4DCT provides more accurate patient-specific ITVs.,,, Therefore, in this study, the ITV resulted by 4DCT was considered as a reference against which all other ITVs were compared. The 4DCT acquisition was carried out in a helical mode with gantry rotation speed of 0.5 s/rotation and pitch factor of 0.1. Real-time position management™ system (Varian Medical Systems, Palo Alto, CA, USA) was used to generate the respiratory wave of the moving targets, and this wave was used for retrospective reconstruction of 4DCT images. 4DCT resulted in ten 3DCT data sets corresponding to different phases of respiratory motion.
For each respiratory motion, full-fan (FF) and half-fan (HF) CBCT scans were also acquired (slice thickness 2 mm) using Varian On-Board Imager system attached with Clinac2100CD Linear Accelerator (Varian Medical Systems, Palo Alto, CA, USA). The experimental setup for CBCT acquisition is shown in [Figure 1].
All the CT images were imported into Eclipse™ treatment planning system (Varian Medical Systems, Palo Alto, CA, USA) and co-registered based on the pixel data using rigid registration algorithm (Version 8.6.15). The volume of the target ball was autosegmented in each image by using the window level W = −715; L = +1000 for average CT, W = −915; L= +1000 for CBCT and W = −650; L = +1000 for all remaining CT images. Since target boundary blurring was different for various scans, different window level was used. The 4DCT ITVs were generated by merging all 10 phase target volumes, whereas the BH-ITV was generated by boolean of target volumes of CT phases 0%, 50%, and fast CT.
The absolute volume of each ITV was measured, and the percentage deviations were calculated with respect to 4DCT ITVs. ITVs of all CT techniques were statistically compared with ITVs of 4DCT using the Student's t-test (two-tailed), and the differences were considered to be significant for P < 0.05. Intersection/overlapping the volume of each ITV with 4DCT ITV was determined. The center of mass coordinates was also determined for each ITVs.
| > Results|| |
[Table 2] shows the ITVs resulted by various CT techniques for all ten cases. ITVs determined by fast CT, slow CT, BH-CT, FF-CBCT, and HF-CBCT were all considerably smaller than the 4DCT ITVs, and the difference was statistically significant (P < 0.05). On the contrary, the average CT did not show any difference (P = 0.17).
|Table 2: Internal target volume resulted by various computed tomography techniques|
Click here to view
Fast CT had the maximum deviation from 4DCT ITVs, and the FF-CBCT scans were slightly better over slow CT. Whereas HF-CBCT scans and BH-CT scans resulted in less deviation. The average percentage deviations for fast CT, slow CT, FF-CBCT, HF-CBCT, and BH-CT were −46.1% ± 20.9%, −34.3% ± 11.0%, −26.3% ± 8.7%, −12.7% ± 4.7%, and −11.1% ± 8.5% respectively, whereas the same for average CT was only −4.7% ± 9.8%.
[Figure 2] shows the intersection volumes of various CT techniques. The intersecting volumes of all CT scans were almost equal to the actual ITVs of each CT technique and this indicated that the ITVs of all CT scans were inside the 4DCT ITVs.
|Figure 2: Comparison of intersection/overlapping volumes (with four-dimensional computed tomography) of different computed tomography techniques with respective individual internal target volumes|
Click here to view
Center of mass shifts of all ITVs was calculated with respect to 4DCT ITV. [Table 3] shows the center of mass shifts of various ITVs. The maximum shift was observed for fast CT; on the other hand, the minimum shift was observed for BH-CT. The 3D displacement vectors were also calculated. On average, 3D displacement vectors for fast CT, slow CT, FF-CBCT, HF-CBCT, BH-CT, and average CT were 3.6 ± 2.2 mm, 1.2 ± 0.8 mm, 1.0 ± 0.8 mm, 1.1 ± 0.6 mm, 0.5 ± 0.6 mm, and 0.5 ± 0.5 mm, respectively.
|Table 3: Center of mass shifts of internal target volumes of various computed tomography with 4 dimensional computed tomography internal target volumes|
Click here to view
| > Discussion|| |
In this study, the accuracy of different CT scan techniques such as fast CT, slow CT, BH-CT, FF-CBCT, HF-CBCT, and average CT were evaluated for delineating the ITV by comparing these volumes with ITVs of 4DCT. ITVs of all CT techniques were smaller than 4DCT ITVs. Among all CT techniques, maximum deviations were observed for fast CT followed by slow CT and FF-CBCT. Both BH-CT and HF-CBCT resulted less deviation (<13%). The number of respiratory cycles in a given CT acquisition would be more for HF-CBCT compared to other CT techniques, i.e., FF-CBCT, slow CT, and fast CT. Therefore, the HF-CBCT showed less ITV deviation with 4DCT. On the other hand, the BH-CT and average CT were derived from 4DCT; hence, they were superior to HF-CBCT.
From the [Table 1], it can be seen that the cases 9 and 10 were having relatively less complicated respiratory motion (amplitude <5 mm in all axis) and this resulted in less ITV deviation [Table 2]. On the other hand, cases 1, 2, and 3 were having maximum ITV deviation because they had complicated respiratory motion. This indicated that the ITV deviation depends on the complexity of respiratory motion.
One of the purposes of this work was to find out an alternative CT technique for delineating ITV in the absence of 4DCT. As discussed earlier, the BH-CT and average CT used in this study were derived from the 4DCT; so they cannot be an alternative. Therefore, HF-CBCT may be used as an alternative to 4DCT. However, HF-CBCT showed a small deviation (~13%) with 4DCT; therefore, an additional margin needs to be added with HF-CBCT ITV. For our cases, we measured the additional margin required for the ITVs of HF-CBCT to be identical with 4DCT ITVs. The [Table 4] shows these margins in each six directions measured for ten cases. On average, the additional margin required for HF-CBCT was cranial: 2.8 ± 1.5 mm; caudal: 1.8 ± 0.8 mm; anterior: 1.5 ± 0.9 mm; posterior: 1.2 ± 0.6 mm; left: 2.3 ± 0.9 mm and right: 2.2 ± 0.7 mm. However, one should be aware that these margins are patient-specific and depend on the respiratory motion.
|Table 4: Extra margins required for internal target volumes of half-fan cone beam computed tomography to be equal with four-dimensional computed tomography internal target volumes|
Click here to view
Delineating the tumor boundary in CBCT would be intricate due to its poor image quality compared to 4DCT images. However, we did not encounter this problem because we used a phantom; however, in real patients, it may be an issue. Hence, CBCT-based ITV method may be best suitable for tumors surrounded entirely by lung tissues and not for tumors close to mediastinum and chest-wall. In addition, this study included single target size (3 cm diameter), whereas different target size may yield different ITV deviation.
Another important finding of this study was that the center of mass shift of fast CT scan was observed to be maximum [Table 3], whereas for all other CT techniques, the center of mass shifts was <1 mm. This indicated that population based margin in fast CT should to be used with caution. Because the fast CT target volumes may not represent the mean target position. Our results agree with the previous investigations., Wang et al. showed the feasibility of verifying the ITV of 4DCT with 3DCBCT in a phantom experiment. The ITV deviation between CBCT and 4DCT was 8.7%, and localization accuracy was within 1 mm. Similarly, in a recent study, Wang et al. also compared the 3DCBCT with 4DCT maximum intensity projection (MIP) for ITV localization and found that the 4DCT MIP was generally larger than (10% - ITV segmentation based on gradient method) those obtained with CBCT. They also observed a maximum difference in the centroid position was <1.4 mm between two modalities. While we observed a mean deviation of 13% and the center of mass shift was <1.0 mm.
The following aspects need to be considered when interpreting the results. In this study, 4DCT was considered as the golden standard, and it is a rational approach. However, the accuracy of 4DCT may be reduced in very slow and rapid respiration. The slow CT technique used in this study does not represent a typical slow CT (ideally scanning time 4 s per slice should be used)., Since gantry rotation time could not be increased more than 1 s; the ideal slow CT was not possible in our CT scanner. As a result, the ITVs resulted by our slow CT may be smaller than ITVs of typical slow CT scan.
Further, Wang et al. showed that the ITV resulted by voluntary BH-CT was larger (24.6% larger on average) than 4DCT ITV. In contrast, we observed consistently lesser ITVs for BH-CT; because in our study, the BH-CT was derived from the 0% to 50% phases of 4DCT. In addition, the voluntary BH-CT would have better image quality and less residual motion compared to the BH-CT derived from the 4DCT. Finally, all investigations were carried out for regular breathing pattern, and a separate study is required for irregular breathing.
| > Conclusions|| |
In this study, different CT and CBCT techniques were compared with 4DCT for delineating the ITVs. All these CT techniques underestimated the ITV, and the difference was statistically significant except for average CT. In the absence of 4DCT, the HF-CBCT target volumes with appropriate margin may be a reasonable approach for deriving the ITV.
We would acknowledge the Department of Science and Technology - Grant No: IR/SO/LS 02/2003, Government of India.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Seppenwoolde Y, Shirato H, Kitamura K, Shimizu S, van Herk M, Lebesque JV, et al.
Precise and real-time measurement of 3D tumor motion in lung due to breathing and heartbeat, measured during radiotherapy. Int J Radiat Oncol Biol Phys 2002;53:822-34.
Maria DK, Agarwal A, Yoganathan SA, Gowtham Raj D, Kumar S. Investigation of Inter and Intra-Fractional Uncertainties in Lung IMRT Delivery. World Congress on Medical Physics and Biomedical Engineering; 2012. p. 1857-9.
Yoganathan SA, Maria Das KJ, Kumar S. Computational model to simulate the interplay effect in dynamic IMRT delivery. J Phys Conf Ser 2014;489:12042-7.
Rowell NP, Williams CJ. Radical radiotherapy for stage I/II non-small cell lung cancer in patients not sufficiently fit for or declining surgery (medically inoperable): a systematic review. Thorax. 2001;56:628–38.
Qiao X, Tullgren O, Lax I, Sirzén F, Lewensohn R. The role of radiotherapy in treatment of stage I non-small cell lung cancer. Lung Cancer 2003;41:1-11.
Kong FM, Ten Haken RK, Schipper MJ, Sullivan MA, Chen M, Lopez C, et al.
High-dose radiation improved local tumor control and overall survival in patients with inoperable/unresectable non-small-cell lung cancer: Long-term results of a radiation dose escalation study. Int J Radiat Oncol Biol Phys 2005;63:324-33.
Terakedis B, Sause W. Radiation dose escalation in stage III non-small-cell lung cancer. Front Oncol 2011;1:47.
Nyman J, Johansson KA, Hultén U. Stereotactic hypofractionated radiotherapy for Stage I non-small cell lung cancer – Mature results for medically inoperable patients. Lung Cancer 2006;51:97-103.
Salama JK, Stinchcombe TE, Gu L, Wang X, Morano K, Bogart JA, et al.
Pulmonary toxicity in stage III non-small cell lung cancer patients treated with high-dose (74 Gy) 3-dimensional conformal thoracic radiotherapy and concurrent chemotherapy following induction chemotherapy: A secondary analysis of Cancer and Leukemia Group B (CALGB) trial 30105. Int J Radiat Oncol Biol Phys 2011;81:e269-74.
Underberg RW, Lagerwaard FJ, Slotman BJ, Cuijpers JP, Senan S. Use of maximum intensity projections (MIP) for target volume generation in 4DCT scans for lung cancer. Int J Radiat Oncol Biol Phys 2005;63:253-60.
Wang Z, Wu QJ, Marks LB, Larrier N, Yin FF. Cone-beam CT localization of internal target volumes for stereotactic body radiotherapy of lung lesions. Int J Radiat Oncol Biol Phys 2007;69:1618-24.
Wang L, Chen X, Lin MH, Xue J, Lin T, Fan J, et al.
Evaluation of the cone beam CT for internal target volume localization in lung stereotactic radiotherapy in comparison with 4D MIP images. Med Phys 2013;40:111709.
Sixel KE, Ruschin M, Tirona R, Cheung PC. Digital fluoroscopy to quantify lung tumor motion: Potential for patient-specific planning target volumes. Int J Radiat Oncol Biol Phys 2003;57:717-23.
Lagerwaard FJ, Van Sornsen de Koste JR, Nijssen-Visser MR, Schuchhard-Schipper RH, Oei SS, Munne A, et al.
Multiple “slow” CT scans for incorporating lung tumor mobility in radiotherapy planning. Int J Radiat Oncol Biol Phys 2001;51:932-7.
Wurstbauer K, Deutschmann H, Kopp P, Sedlmayer F. Radiotherapy planning for lung cancer: Slow CTs allow the drawing of tighter margins. Radiother Oncol 2005;75:165-70.
Vergalasova I, Maurer J, Yin FF. Potential underestimation of the internal target volume (ITV) from free-breathing CBCT. Med Phys 2011;38:4689-99.
Wong JW, Sharpe MB, Jaffray DA, Kini VR, Robertson JM, Stromberg JS, et al.
The use of active breathing control (ABC) to reduce margin for breathing motion. Int J Radiat Oncol Biol Phys 1999;44:911-9.
Mah D, Hanley J, Rosenzweig KE, Yorke E, Braban L, Ling CC, et al.
Technical aspects of the deep inspiration breath-hold technique in the treatment of thoracic cancer. Int J Radiat Oncol Biol Phys 2000;48:1175-85.
Ohara K, Okumura T, Akisada M, Inada T, Mori T, Yokota H, et al.
Irradiation synchronized with respiration gate. Int J Radiat Oncol Biol Phys 1989;17:853-7.
Kubo HD, Hill BC. Respiration gated radiotherapy treatment: A technical study. Phys Med Biol 1996;41:83-91.
Yoganathan SA, Das KJ, Raj DG, Kumar S. Dosimetric verification of gated delivery of electron beams using a 2D ion chamber array. J Med Phys 2015;40:68-73.
] [Full text]
D'Souza WD, Naqvi SA, Yu CX. Real-time intra-fraction-motion tracking using the treatment couch: A feasibility study. Phys Med Biol 2005;50:4021-33.
Schweikard A, Glosser G, Bodduluri M, Murphy MJ, Adler JR. Robotic motion compensation for respiratory movement during radiosurgery. Comput Aided Surg 2000;5:263-77.
Keall PJ, Kini VR, Vedam SS, Mohan R. Motion adaptive x-ray therapy: A feasibility study. Phys Med Biol 2001;46:1-10.
Yoganathan SA, Maria Das KJ, Agarwal A, Senthil Kumar SK, Velmurugan J, Kumar S. Synchronization of Intra-Fractional Motion in Dynamic IMRT Delivery. World Congress on Medical Physics and Biomedical Engineering; 2013. p. 1923-5.
Yoganathan SA, Maria Das KJ, Agarwal A, Kumar S. Performance evaluation of respiratory motion-synchronized dynamic IMRT delivery. J Appl Clin Med Phys 2013;14:4103.
Keall PJ, Mageras GS, Balter JM, Emery RS, Forster KM, Jiang SB, et al.
The management of respiratory motion in radiation oncology report of AAPM Task Group 76. Med Phys 2006;33:3874-900.
International Commission on Radiation Units and Measurements. Prescribing, Recording and Reporting Photon Beam Therapy. Supplement to Report 50. Report 62; 1999.
Underberg, René WM, Sörnsen de Koste JR, Lagerwaard FJ, Vincent A, Slotman BJ, et al
. A dosimetric analysis of respiration-gated radiotherapy in patients with stage III lung cancer. Radiat Oncol 2006;1.
Underberg RW, Lagerwaard FJ, Slotman BJ, Cuijpers JP, Senan S. Benefit of respiration-gated stereotactic radiotherapy for stage I lung cancer: An analysis of 4DCT datasets. Int J Radiat Oncol Biol Phys 2005;62:554-60.
Dinkel J, Welzel T, Bolte H, Hoffmann B, Thierfelder C, Mende U, et al.
Four-dimensional multislice helical CT of the lung: Qualitative comparison of retrospectively gated and static images in an ex-vivo
system. Radiother Oncol 2007;85:215-22.
Guckenberger M, Wilbert J, Meyer J, Baier K, Richter A, Flentje M. Is a single respiratory correlated 4D-CT study sufficient for evaluation of breathing motion? Int J Radiat Oncol Biol Phys 2007;67:1352-9.
Wang L, Hayes S, Paskalev K, Jin L, Buyyounouski MK, Ma CC, et al.
Dosimetric comparison of stereotactic body radiotherapy using 4D CT and multiphase CT images for treatment planning of lung cancer: Evaluation of the impact on daily dose coverage. Radiother Oncol 2009;91:314-24.
Chinneck CD, McJury M, Hounsell AR. The potential for undertaking slow CT using a modern CT scanner. Br J Radiol 2010;83:687-93.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]