Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 
Home Print this page Email this page Small font size Default font size Increase font size Users Online: 466


 
 Table of Contents 
ORIGINAL ARTICLE
Year : 2021  |  Volume : 44  |  Issue : 2  |  Page : 110-115  

Dosimetric evaluation of analytic anisotropic algorithm and Acuros XB algorithm using in-house developed heterogeneous thorax phantom and homogeneous slab phantom for stereotactic body radiation therapy technique


School of Studies in Physics, Vikram University, Ujjain, Madhya Pradesh, India

Date of Submission07-Oct-2020
Date of Decision11-May-2021
Date of Acceptance16-May-2021
Date of Web Publication23-Oct-2021

Correspondence Address:
Priyusha Bagdare
School of Studies in Physics, Vikram University, Ujjain - 456 001, Madhya Pradesh
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/rpe.rpe_52_20

Rights and Permissions
  Abstract 


To perform patient-specific quality assurance (QA), the accuracy of the dose calculation algorithm is vital, especially in the lung cancer stereotactic body radiation therapy (SBRT). The present study is based on the evaluation of two widely used algorithms, analytical anisotropic algorithm (AAA) and Acuros XB (AXB) inside the in-house developed heterogeneous thorax phantom (HTP) and a homogeneous slab phantom (HSP) simultaneously. To evaluate dosimetric differences between the two algorithms, point dose measurement was performed for pretreatment QA plans of 35 lung cancer patients by keeping the same monitor units and beam angles as those for the actual patient treatment. The dose was calculated on the Eclipse treatment planning system inside both the medium by using both AAA and AXB algorithms. Plans were delivered on the Edge linear accelerator (LA) (Varian Medical Systems, Palo Alto, CA, USA), and measurements were taken by using a 0.01 cc ion chamber and DOSE1 electrometer. Statistical analysis was performed on the observed data set, and percentage (%) variations between the measured and planned doses were calculated and analyzed. The mean % variations between the measured and planned doses inside HTP for all QA plans were found to be 2.61 (standard deviation [SD]: 0.66) and 2.19 (SD: 0.64) for AAA and AXB algorithms, respectively. Whereas, inside HSP, it was found to be 1.79 (SD: 0.74) and 1.64 (SD: 0.70) for AAA and AXB algorithms, respectively. The mean % difference between the measured dose and the planned dose was derived to be statistically significant for HTP, however, it was found to be statistically insignificant inside the HSP at P < 0.01. The Pearson's correlation coefficient test showed a strong positive correlation between the measured dose and the planned dose for both AAA and AXB inside HTP as well for HSP. The results obtained from this study showed that as the actual patient body is heterogeneous, thus to get more realistic results, patient-specific QA must be performed inside the heterogeneous phantom instead of homogeneous. Moreover, in the homogeneous medium, both the algorithms predict the dose efficiently, however, in heterogeneous medium, AAA over/under predicts the dose, whereas AXB shows good concurrence with measurements.

Keywords: Acuros XB, analytical anisotropic algorithm, heterogeneous thorax phantom, homogeneous slab phantom, patient-specific quality assurance


How to cite this article:
Dubey S, Bagdare P, Ghosh S. Dosimetric evaluation of analytic anisotropic algorithm and Acuros XB algorithm using in-house developed heterogeneous thorax phantom and homogeneous slab phantom for stereotactic body radiation therapy technique. Radiat Prot Environ 2021;44:110-5

How to cite this URL:
Dubey S, Bagdare P, Ghosh S. Dosimetric evaluation of analytic anisotropic algorithm and Acuros XB algorithm using in-house developed heterogeneous thorax phantom and homogeneous slab phantom for stereotactic body radiation therapy technique. Radiat Prot Environ [serial online] 2021 [cited 2021 Dec 8];44:110-5. Available from: https://www.rpe.org.in/text.asp?2021/44/2/110/329139




  Introduction Top


Stereotactic body radiation therapy (SBRT) has become a unique treatment technique for patients with inoperable or early-stage lung cancer.[1],[2] The main attribute that separates SBRT from conventional radiotherapy is the delivery of accelerated doses in a few fractions. In SBRT instead of administrating conventional fractionation (2–3 Gy/fraction), hypofractionation (>6 Gy/fraction) typically delivered in 3–5 fractions is given to the tumor volume.[3] To get the maximum benefit of this technique, very tight margins are set between the tumor and organ at risk (OAR) during the contouring; because of this, a sharp dose gradient gets formed between the two while treatment delivery. Therefore, a small error during the entire treatment process may risk either underdose to tumor or overdose to OARs. Thus, a proper quality assurance (QA) procedure is very important for reducing the chances of errors in SBRT. Among all the QA, patient-specific QA is a vital component of the entire plan delivery procedure as it provides the confidence that the delivered dose distribution will be accurately matched with the planned dose distribution.[4],[5] As per the International Commission on Radiation Units and Measurements 83 (ICRU 83), the variation in dosimetric results should not be more than ±3%, however, focus is always there to make it within ±1%, especially for SBRT cases.[6] Many factors lead to a discrepancy in the outcome of patient-specific QA; dosimetric phantom and the type of algorithm used for dose calculation are the two important factors in between them. Phantoms that are regularly utilized for patient-specific QA are homogeneous and have a simple geometry. As a result, for any algorithm to calculate dose within such a simple geometry is quite easy.[7] However, in actual, the human body is inhomogeneous in nature, and hence, if a heterogeneous phantom is used for plan verification, then it is indeed challenging for any algorithm to calculate the accurate and precise dose inside it.[8] To properly account for the in-homogeneities, high-end algorithms that employ heterogeneity correction are required for the exact dose calculation.[9] However, there exists considerable variability in dose calculation algorithms which are commercially available with the treatment planning systems (TPSs). In general, these algorithms possess either accuracy or computation speed but not both. Most of the commercial TPSs use a standard algorithm that possesses fast calculation speed and approximates the dose inside the inhomogeneous medium such as the thorax region. Although the calculations are quick, the main disadvantage of these algorithms is that they are not consistent in predicting doses, especially when heterogeneity is involved.[10],[11] Among the widely used algorithms, namely pencil beam convolution, collapsed cone convolution, and analytical anisotropic algorithm (AAA), AAA shows better accuracy in lung cancer treatment, but it underestimates/overestimates the dose at the boundary region depending upon the density involved, such as lung/soft tissue or bone/soft tissue.[12],[13] Dose calculation algorithms based on Monte Carlo (MC) codes have shown promising implications in lung treatment plans.[14],[15] However, the main disadvantage of these algorithms is the long computation time which makes them unsuitable for routine clinical applications. Acuros XB (AXB) introduced by the Eclipse TPS by Varian Medical Systems (Palo Alto, USA), similar to classic MC methods, shows the same accuracy as by MC calculations in heterogeneous media with smaller calculation time. Several studies have reported the better dose prediction accuracy of the AXB over AAA inside the heterogeneous medium when compared with the measurements and MC simulations.[16],[17],[18]

However, to our knowledge, no study has been reported in the literature to examine the performance of these algorithms using heterogeneous and homogeneous phantom simultaneously for SBRT technique. The present study deals with the pretreatment absolute point dose verification of 35 patients treated with SBRT technique using AAA and AXB algorithms inside the in-house developed heterogeneous thorax phantom (HTP) and homogeneous slab phantom (HSP).


  Materials and Methods Top


Two phantoms were taken for the study, namely HTP and HSP. The HTP was built from materials with a similar density as that of the actual human thorax parts involved.[19] Materials used are of organic origin to meet the bioneeds of clinical investigations. The thorax phantom was divided into the lung, ribs, and soft-tissue regions. The lung prototype was made with porous sawdust of pinewood with a mean density of 0.24 g/cc. The rib cage was made from a material with the same density as that of rib bones with a mean density of 1.84 g/cc. Natural honeybee's wax was used to provide a coating on the rib cage for replicating the tissue-equivalent material with a mean density of 0.86 g/cc, as shown in [Figure 1].
Figure 1: In-house developed heterogeneous thorax phantom

Click here to view


However, to design the HSP, an arrangement of 15 SP34 slabs (IBA Dosimetry GmbH, Schwarzenbruck, Germany) of a thickness of 1 cm made up of polystyrene C8H8 (composition: 98% polystyrene +2% TiO2) was used. Each slab was of uniform density with an effective atomic number of 5.74, mean density 0.99 g/cc, and area of 30 cm × 30 cm.

Computed tomography (CT) images of the designed HTP, as well as that of the HSP with 3 mm slice thickness, were taken on SOMATOM SCOPE Power (Siemens Medical Systems, Germany). The acquired CT images of the phantom were then imported into the TPS Eclipse version 13.6 (Varian Medical Systems, Palo Alto, CA, USA).

Volumetric-modulated arc therapy or intensity-modulated radiotherapy treatment plans were generated for the selected patients with the prescribed dose of 8 Gy per fraction. Using photon optimizer, plans were optimized keeping the same optimization parameters and monitor units for both the algorithms. The final dose calculation was performed on TPS by using AAA (vs. 13.6.23) and AXB (vs. 13.6.23) algorithms with the grid size taken as 0.25 cm. Two or three arcs and/or 5–7 coplanar/noncoplanar fields using 10 MV flattening filter-free (FFF) photon energy were used to create these plans. To carry out absolute point dosimetry, the pretreatment QA plans for all the patients were recomputed in the CT images of HTP and HSP using both the algorithms, by keeping the same gantry angle as that was there in the actual plan. After approving all the QA plans, they were scheduled on time planner for delivery. Plans were executed on Edge linear accelerator (LA) (Varian Medical Systems, Palo Alto, CA) which is equipped with high-definition multileaf collimator, couch with 6° of freedom, and jaw tracking technology for a reduction in leakage and out-of-field dose along with FFF beams for facilitating faster treatment.[20] Measurement of absolute point dose at a reference point on LA was carried out inside the HTP and HSP by using a 0.01 cc ion chamber and DOSE1 electrometer. To verify the position of the ion chamber inside the phantoms, primary CT images obtained using the Eclipse TPS were matched and verified by taking cone-beam CT following the standard imaging procedure. The percentage variation between the planned TPS dose and the measured dose on LA was calculated using the following relation:

Percentage variation = ([Planned dose − measured dose]/measured dose) ×100.

Descriptive statistics were used in the form of mean and standard deviation (SD) to assess the observed data. Statistical analysis of the data was performed employing MS Excel and SPSS version 22 (IBM SPSS Statistics for Windows, Armonk, NY: IBM Corp). Aiming to analyze the correlation between the measured dose on LA and the planned dose by TPS, Pearson's correlation test was applied. The normality of all data was screened using the Kolmogorov–Smirnov test. A statistically significant difference between each set of dosimetric parameters was calculated by paired sample t-test, with P < 0.01 being statistically significant.


  Results Top


The comparison of the measured dose and planned dose by AAA and AXB for 35 patients QA plan inside HTP is summarized in [Table 1]. The mean % variation between the measured and planned doses inside HTP for all QA plans was found to be 2.61 (SD: 0.66) and 2.19 (SD: 0.64) for AAA and AXB algorithms, respectively. The percentage variation between the measured and planned doses was found to be higher in AAA as compared to that of AXB for the HTP. The difference between the % error calculated by two algorithms was found to be statistically significant for HTP. A significance level of 0.01 (P < 0.01) was taken as a reference to analyze the dataset.
Table 1: Comparison of the measured dose and planned dose by analytical anisotropic algorithm and Acuros XB inside heterogeneous thorax phantom

Click here to view


Pearson's correlation coefficient was calculated for the data set, and our findings showed a strong positive linear correlation between the measured dose and the planned dose by both AAA (r = 0.97) and AXB (r = 0.98), respectively, for HTP. However, a stronger positive correlation with measured dose was shown by AXB as compared to that of the AAA.

The comparison of the measured dose and the planned dose for all QA plans using HSP is shown in [Table 2] for AAA and AXB algorithms.
Table 2: Comparison of the measured dose and planned dose by analytical anisotropic algorithm and Acuros XB inside homogeneous slab phantom

Click here to view


The mean % variation between the measured and planned doses was found to be 1.79 (SD: 0.74) and 1.64 (SD: 0.70) for AAA and AXB algorithms, respectively, inside HSP. The difference between the % error calculated by two algorithms was found to be statistically insignificant for HSP with a significance level of 0.01 (P < 0.01). Pearson's correlation coefficient showed the equal and strong positive correlation between the measured dose and the planned dose for both AAA (r = 0.99) and AXB (r = 0.99), respectively.


  Discussions Top


At present, most of the dosimetric practices are based on the guiding principle set by the ICRU 83, AAPM task group report 120, and TRS 398, and the dosimetric phantoms recommended by them for dosimetry are mostly water or water equivalent.[21],[22],[23] The results of this study indicate that the chances of errors are more if the patient-specific absolute point dose measurement was carried out inside the homogeneous medium instead of heterogeneous. Inside any homogeneous medium, dose calculation is quite easy, straightforward, and accurate, thus the error observed between the measured dose and the planned dose will always be less. However, if the medium is heterogeneous, then it requires lots of density corrections, and thus, the calculation of dose is quite tedious, especially at the boundary regions.[24] The present study supports the same connotation as less variation was observed between the measured dose and the planned dose inside the HSP as compared to that of the HTP. However, as we know that the patient body is inhomogeneous, and thus, to get the realistic results of patient-specific dosimetry, it should be performed inside the heterogeneous phantom instead of homogeneous.

Another important conclusion which we have drawn from this study is that if the medium of patient-specific dosimetry is homogeneous, then both AAA and AXB algorithms are equally good for dose calculation. Moreover, the results obtained from both the algorithms come well within the tolerance limit. However, keeping all the parameters identical, if the same measurements are performed inside the heterogeneous medium, then it is observed that in between AAA and AXB, AXB predicts the dose more accurately.[25] AAA has a tendency to overestimate or underestimate the dose inside the heterogeneous medium depending on the boundary region, field size, and position of reference point.[26] In the current study, as the reference point of measurement inside the HTP was kept completely inside the lung region, thus the dose calculated by AAA was observed within the tolerance limits of ±3%. However, if the reference point of measurement is chosen at the boundary of two regions, then this variation may go beyond ±5% as reported by the many authors.[27],[28]

Thus, overall, if we want to evaluate the efficiency of any algorithm, then the measurements must be performed inside the heterogeneous medium. Because, if perform the measurements inside the homogeneous medium, then the dose predicted by both the algorithms is less variant and is very near to the measured dose and thus we cannot judge the correctness of any algorithm accurately.


  Conclusions Top


Various commercial and noncommercial radiotherapy QA phantoms are in use for QA studies, education, training, and research. Commercially available phantoms which are generally used for patient-specific QA are homogeneous. QA performed on these homogeneous phantoms shows less variation, while in an actual situation, this variation can be high if the same measurements are taken inside the heterogeneous phantom. In addition to this, the type of algorithm used for dose calculation plays an important role in patient-specific dosimetric outcomes.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Fukumoto S, Shirato H, Shimzu S, Ogura S, Onimaru R, Kitamura K, et al. Small-volume image-guided radiotherapy using hypofractionated, coplanar, and noncoplanar multiple fields for patients with inoperable Stage I nonsmall cell lung carcinomas. Cancer 2002;95:1546-53.  Back to cited text no. 1
    
2.
Sibley GS, Jamieson TA, Marks LB, Anscher MS, Prosnitz LR. Radiotherapy alone for medically inoperable Stage I non-small-cell lung cancer: The Duke experience. Int J Radiat Oncol Biol Phys 1998;40:149-54.  Back to cited text no. 2
    
3.
Benedict SH, Yenice KM, Followill D, Galvin JM, Hinson W, Kavanagh B, et al. Stereotactic body radiation therapy: The report of AAPM Task Group 101. Med Phys 2010;37:4078-101.  Back to cited text no. 3
    
4.
Galvin JM, Bednarz G. Quality assurance procedures for stereotactic body radiation therapy. Int J Radiat Oncol Biol Phys 2008;71:S122-5.  Back to cited text no. 4
    
5.
Potters L, Kavanagh B, Galvin JM, Hevezi JM, Janjan NA, Larson DA, et al. American Society for Therapeutic Radiology and Oncology (ASTRO) and American College of Radiology (ACR) practice guideline for the performance of stereotactic body radiation therapy. Int J Radiat Oncol Biol Phys 2010;76:326-32.  Back to cited text no. 5
    
6.
Gregoire V, Mackie R. State of the art on dose prescription, reporting and recording in Intensity-Modulated Radiation Therapy (ICRU report No. 83). Cancer Radiotherapie 2011;15:555-9.  Back to cited text no. 6
    
7.
Gurjar OP, Mishra SP, Bhandari V, Pathak P, Patel P, Shrivastav G. Radiation dose verification using real tissue phantom in modern radiotherapy techniques. J Med Phys 2014;39:44-9.  Back to cited text no. 7
[PUBMED]  [Full text]  
8.
Ding W, Johnston PN, Wong TP, Bubb IF. Investigation of photon beam models in heterogeneous media of modern radiotherapy. Australas Phys Eng Sci Med 2004;27:39-48.  Back to cited text no. 8
    
9.
Kroon PS, Hol S, Essers M. Dosimetric accuracy and clinical quality of Acuros XB and AAA dose calculation algorithm for stereotactic and conventional lung volumetric modulated arc therapy plans. Radiat Oncol 2013;8:149.  Back to cited text no. 9
    
10.
Gray A, Oliver LD, Johnston PN. The accuracy of the pencil beam convolution and anisotropic analytical algorithms in predicting the dose effects due to attenuation from immobilization devices and large air gaps. Med Phys 2009;36:3181-91.  Back to cited text no. 10
    
11.
Lu L, Goma G, Wang J, Gupta N, Huang Z, Lo S, et al. A practical method to evaluate and verify dose calculation algorithms in the treatment planning system of radiation therapy. Int J Med Phys Clin Eng Radiat Oncology 2013;2:76-87.  Back to cited text no. 11
    
12.
Van Esch A, Tillikainen L, Pyykkonen J, Tenhunen M, Helminen H, Siljam Aki S, et al. Testing of the analytical anisotropic algorithm for photon dose calculation. Med Phys 2006;33:4130-48.  Back to cited text no. 12
    
13.
Robinson D. Inhomogeneity correction and the analytic anisotropic algorithm. J Appl Clin Med Phys 2008;9:112-22.  Back to cited text no. 13
    
14.
Krieger T, Sauer OA. Monte Carlo- versus pencil-beam-/collapsed-cone-dose calculation in a heterogeneous multi-layer phantom. Phys Med Biol 2005;50:859-68.  Back to cited text no. 14
    
15.
Panettieri V, Wennberg B, Gagliardi G, Duch MA, Ginjaume M, Lax I. SBRT of lung tumours: Monte Carlo simulation with PENELOPE of dose distributions including respiratory motion and comparison with different treatment planning systems. Phys Med Biol 2007;52:4265-81.  Back to cited text no. 15
    
16.
Huang B, Wu L, Lin P, Chen C. Dose calculation of Acuros XB and Anisotropic Analytical Algorithm in lung stereotactic body radiotherapy treatment with flattening filter free beams and the potential role of calculation grid size. Radiat Oncol 2015;10:53.  Back to cited text no. 16
    
17.
Bush K, Gagne IM, Zavgorodni S, Ansbacher W, Beckham W. Dosimetric validation of Acuros XB with Monte Carlo methods for photon dose calculations. Med Phys 2011;38:2208-21.  Back to cited text no. 17
    
18.
Rana S. Clinical dosimetric impact of acuros XB and analytical anisotropic algorithm (AAA) on real lung cancer treatment plans: Review. Int J Cancer Ther Oncol 2014;2:02019.  Back to cited text no. 18
    
19.
Dubey S, Ghosh S, Bagdare P. Radiological properties of high energy X-rays inside the in house developed heterogeneous thorax phantom. AIP Conf Proce 2020;2224:6-10.  Back to cited text no. 19
    
20.
Wen N, Li H, Song K, Chin-Snyder K, Qin Y, Kim J, et al. Characteristics of a novel treatment system for linear accelerator-based stereotactic radiosurgery. J Appl Clin Med Phys 2015;16:125-48.  Back to cited text no. 20
    
21.
Hodapp N. The ICRU Report 83: Prescribing, recording and reporting photon-beam intensity-modulated radiation therapy (IMRT). Strahlenther Onkol 2012;188:97-9.  Back to cited text no. 21
    
22.
Eaton D, Bass G, Booker P, Byrne J, Duane S, Frame J, et al. IPEM code of practice for high-energy photon therapy dosimetry based on the NPL absorbed dose calibration service. Phys Med Biol 2020;65:1950-06.  Back to cited text no. 22
    
23.
Low DA, Moran JM, Dempsey JF, Dong L, Oldham M. Dosimetry tools and techniques for IMRT. Med Phys 2011;38:1313-38.  Back to cited text no. 23
    
24.
Followill DS, Evans DR, Cherry C, Molineu A, Fisher G, William F, et al. Design, development, and implementation of the radiological physics center's pelvis and thorax anthropomorphic quality assurance phantoms. Med Phys 2007;34:2070-76.  Back to cited text no. 24
    
25.
Kathirvel M, Subramanian S, Clivio A, Arun G, Fogliata A, Nicolini G, et al. Critical appraisal of the accuracy of Acuros-XB and Anisotropic Analytical Algorithm compared to measurement and calculations with the compass system in the delivery of RapidArc clinical plans. Radiat Oncol 2013;8:140.  Back to cited text no. 25
    
26.
Alghamdi S, Tajaldeen A. Evaluation of dose calculation algorithms using different density materials for in-field and out-of-field conditions. Exp Oncol 2019;41:46-52.  Back to cited text no. 26
    
27.
Gagné IM, Zavgorodni S. Evaluation of the analytical anisotropic algorithm in an extreme water-lung interface phantom using Monte Carlo dose calculations. J Appl Clin Med Phys 2006;8:33-46.  Back to cited text no. 27
    
28.
Chopra KL, Leo P, Kabat C, Rai DV, Avadhani JS, Kehwar TS, et al. Evaluation of dose calculation accuracy of treatment planning systems in the presence of tissue heterogeneities. Ther Radiol Oncol 2018;2:28-41.  Back to cited text no. 28
    


    Figures

  [Figure 1]
 
 
    Tables

  [Table 1], [Table 2]



 

Top
   
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussions
Conclusions
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed104    
    Printed0    
    Emailed0    
    PDF Downloaded22    
    Comments [Add]    

Recommend this journal