|
 |
| ORIGINAL ARTICLE |
|
| Year : 2016 | Volume
: 39
| Issue : 3 | Page : 138-145 |
|
|
Validation of intensity-modulated radiotherapy commissioning as per recommendations in test plans of the American Association of Physicists in Medicine task group 119 report
Sandeep Kaushik1, Atul Tyagi2, Lalit Kumar3, Man Pal Singh4, Rajender Singh Kundu1, Rajesh Punia1
1 Department of Applied Physics, Guru Jambheshwar University of Science and Technology, Hisar, Haryana, India
2 Department of Radiation Oncology, Dr. B. L. Kapur Memorial Hospital, New Delhi, India
3 Department of Radiation Oncology, RGCI Hospital, New Delhi, India
4 Department of Physics, MMH College, Ghaziabad, Uttar Pradesh, India
| Date of Web Publication | 30-Nov-2016 |
Correspondence Address:
Sandeep Kaushik
Department of Applied Physics, Guru Jambheshwar University of Science and Technology, Hisar - 125 001, Haryana
India
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/0972-0464.194960
In the present study, intensity-modulated radiotherapy (IMRT) commissioning has been validated as per the task group report 119 (TG119) of the American Association of Physicists in Medicine (AAPM). The plans have been done on structure and computed tomography scanned data set downloaded from the AAPM website. IMRT test planning has been performed to achieve conformed dose and dose distribution similar to that described in the AAPM TG119 report. Point dose measurements with ionization chamber have been taken on solid water phantom at a depth of 7.5 cm. Measurements were performed in two locations, i.e. at target volume and low dose avoidance structure, with planned machine parameters. Gamma evaluation of dose distribution produced by each field has also been done individually using electronic portal imaging device (EPID). Overall mean deviation obtained for ion chamber measurement was −0.012 (standard deviation [SD]: 0.015) with confidence limit (CL) 0.043 and −0.0012 (SD: 0.009) with CL 0.020 in high-dose region and low-dose region, respectively. Overall mean gamma passing in portal dosimetry calculated specifically for EPID has been observed 99.5% with SD 0.33 and CL 1.13. The results obtained for ion chamber dosimetry and portal dosimetry are much better than those reported in the AAPM TG119 report. Better results in gamma evaluation further reduced CL for EPID users. Keywords: Commissioning, intensity-modulated radiotherapy, quality assurance, task group 119
How to cite this article:
Kaushik S, Tyagi A, Kumar L, Singh MP, Kundu RS, Punia R. Validation of intensity-modulated radiotherapy commissioning as per recommendations in test plans of the American Association of Physicists in Medicine task group 119 report. Radiat Prot Environ 2016;39:138-45 |
How to cite this URL:
Kaushik S, Tyagi A, Kumar L, Singh MP, Kundu RS, Punia R. Validation of intensity-modulated radiotherapy commissioning as per recommendations in test plans of the American Association of Physicists in Medicine task group 119 report. Radiat Prot Environ [serial online] 2016 [cited 2023 May 30];39:138-45. Available from: https://www.rpe.org.in/text.asp?2016/39/3/138/194960 |
| Introduction | |  |
Intensity-modulated radiation therapy (IMRT) is extensively used in modern radiotherapy. IMRT is a complex technique, but it plays a vital role in treating cancer patients.[1] In this technique, a very high dose can be delivered to a tumor without much affecting the functionality of surrounding normal tissues and organs.[2] The conformity of dose to target is unremarkable with this technological advancement as compared to conventional method.[3],[4],[5] Commissioning of IMRT needs rigorous measurements, quality assurance (QA), and acceptance. Accuracy of dose calculation by treatment planning system (TPS) is an important aspect of QA.[6] Dose delivery needs a very accurate and repeated treatment QA before treatment of a patient. It includes all machine quality check and TPS QA. There are a number of reports which describe the different QA tests for verification of IMRT plans.[7],[8],[9],[10] Majority of these reports stress on the acceptance test and IMRT commissioning.
In 2008, Radiological Physics Center reported that out of 250 irradiations of head and neck phantom as part of an IMRT verification process, 28% had failed to meet the accuracy criteria of 7% for dose in the low-gradient region and/or 4 mm distance to agreement (DTA) in a high-gradient region.[11] This happened due to poor commissioning and insufficient acceptance between planning and delivery system. To avoid such situations, a number of tests before IMRT delivery, namely, point dose measurement, portal dosimetry, and IMRT pretreatment QA for multileaf collimator leaf positional accuracy have to be performed daily as a routine work. Point dose measurements and portal dose calculation are an accepted testing for IMRT dose distribution. Hence, utilizing such method of testing, a design of comprehensive dosimetric verification is needed to simulate the clinical situation as close as possible in radiotherapy before using IMRT technique to treat a patient. American Association of Physicists in Medicine (AAPM) has provided a measurable confidence limits (CLs) which can be used as baseline value for validation of treatment planning system commissioning for IMRT in the report by task group 119 (TG119) published in 2010.[7] The aim of this study is to validate IMRT commissioning as per test suits and recommendations of the AAPM TG119 report.
| Materials and Methods | | |
The computed tomography (CT) of solid RW3 slab phantom having dimension 30 cm × 30 cm × 15 cm was done on CT scanner (GE Healthcare, UK) taking 0.5 cm slice thickness. CT scan images of phantom and structure set were downloaded from the central server of AAPM website (www.aapm.org) and then, after planning on it, plans were transferred to the local scanned phantom for measurement similar to the one done for IMRT QA. The plans were done on TPS Eclipse version 8.9 (Varian Medical Systems, Palo Alto, CA, USA), in accordance to achieve conformed dose and dose distribution similar to the one described in the AAPM TG119 report. The depth of measurement was taken 7.5 cm in the water equivalent RW3 solid slab phantom. CC13 ionization chamber (IBA dosimetry, Germany) with effective volume 0.125 cc was used for point dose measurement. The planner dose distribution and per field measurement were done by electronic portal imaging device (EPID) mounted on linear accelerator (LINAC) Trilogy Tx (Varian Medical Systems, Palo Alto, CA, USA).
Ion chamber measurements
In all the tests, except multitarget (where all points of measurements are target), measurement was done in two locations: (i) target volume and (ii) low dose avoidance structure. Conversion of chamber reading to dose was done by irradiating the phantom with parallel opposed 10 cm × 10 cm field arranged isocentrically and establishing the ratio of reading to planned dose in that geometry to reduce the daily output variations of LINAC and differences between the phantom, which is water equivalent and actual liquid water. Point dose measurements were done with all fields irradiating the phantom using the planned gantry and collimator angles.[7]
Per field measurements with electronic portal imaging device
Evaluation of the dose distribution produced by individual fields was carried out using the EPID. Dose distributions were analyzed using gamma criteria of 3% dose and 3 mm DTA.[12] The region of interest was defined using threshold dose as 10% of the maximum dose in gamma analysis to cutoff very low-dose regions. Portal dose was also delivered with all fields at planned gantry and collimator angles.[7]
Planning and dosimetric specifications
Five test plans of IMRT were generated with a daily dose of 200 cGy. Planning and calculation was done on TPS using anisotropic analytic algorithm, with a dose rate of 300 monitor units per minute and calculation grid size of 0.25.[13],[14] Six megavoltage (6MV) photon beam was used for all planning. On day 1, simple parallel oppose beam was calculated for 10 cm × 10 cm field size for a dose of 200 cGy to the isocenter, which was taken at 7.5 cm, i.e., at phantom midline. Then, the central dose was measured with an ionization chamber. This measurement was used to set the dose to chamber reading ratio for further tests.[7]
Multitarget test plan
Three cylindrical targets were drawn along the axis of rotation, each having diameter 4.0 cm and length 4.0 cm [Figure 1]a. The central target was made to receive prescribed dose (i.e., 100%). The superior target was made to receive 50% and inferior target 25% of the prescribed dose. Dose to 99% of the volume (D99) and D10 for the three target structures were used to specify the planning goal.[7] Plan was done with seven beams of 6 MV energy at an interval of 50° from the vertical (0°, 50°, 100°, 150°, 310°, 260°, and 210°). Ion chamber measurements were done at isocenter, i.e., at middle of the central target and center of two other targets.[7] | Figure 1: Contour of (a) multitarget test plan, (b) prostate test plan, (c) head and neck test plan, and (d) C shape test plan
Click here to view |
Prostate test plan
[Figure 1]b showing prostate as clinical target volume (CTV) is roughly ellipsoidal with posterior concavity and with right to left (RL), anterior-posterior (AP), and superior-inferior (SI) dimensions of 4.0 cm, 2.6 cm, and 6.5 cm, respectively. The prostate planning target volume (PTV) is expanded 0.6 cm around CTV. The rectum is a cylinder with diameter 1.5 cm that touching the intended posterior aspects of the prostate contour. The PTV includes about one-third of the rectal volume on the widest cross-section of PTV slice. The bladder is roughly ellipsoidal with RL, AP, and SI dimensions of 5.0 cm, 4.0 cm, and 5.0 cm, respectively, and is centered on the superior aspects of the prostate contour. Dose goal for the prostate PTV is specified as D95 and D5 and for rectum and bladder is D30 and D10. The planning dose constraints for different targets are given in [Table 1]. The plan consists of seven fields of 6 MV photon energy at intervals of 50° from vertical (0°, 50°, 100°, 150°, 310°, 260°, and 210°). The ion chamber measurements were carried out at the isocenter, i.e., in mid-PTV for high dose and at 2.5 cm posterior, i.e., at mid-rectum for avoidance structure.[7]
Head and neck test plan
The head and neck PTV [Figure 1]c includes all anterior volumes from the base of the skull to the upper neck, including the posterior neck nodes. The PTV is retracted from the skin by 0.6 cm. There is a gap of about 1.5 cm between the cord and the PTV. The parotid glands are to be avoided and are at the superior aspect of the PTV. Dose goals for head and neck planning are specified as D99, D90, and D20 for PTV and D50 for parotids and maximum dose for the cord. The planning dose constraints for different targets are given in [Table 1]. Plan consists of nine fields of 6 MV photon energy at intervals of 40° from vertical (0°, 40°, 80°, 120°, 160°, 200°, 240°, 280°, and 320°). Ion chamber measurement points were taken at isocenter in mid-PTV and mid-spinal cord, which is 4.0 cm posterior to isocenter.[7]
C shape test plan
The target is a C shape structure that surrounds a central avoidance structure [Figure 1]d. The center core is a cylinder of 1.0 cm radius. The gap between the core and PTV is 0.5 cm; therefore, inner arc of the PTV is 1.5 cm in radius. The outer arc of PTV target is 3.7 cm in radius. The PTV is 8.0 cm long with length of core 10 cm. Two kinds of the problem are given; in the easier type, the central core is to be restricted to 50% of the target dose. In the harder type, the central core is to be restricted to 20% of the target dose. In planning for both easier and harder version, the dose goal was specified as D95 and D10 for C shape PTV, and for central core normal structure, D10 was used as shown in [Table 1]. Nine fields of 6 MV photon energy at intervals of 40° from vertical (0°, 40°, 80°, 120°, 160°, 200°, 240°, 280°, and 320°) were taken for both planning. Ion chamber measurements were taken at central core avoidance and at mid PTV, 2.5 cm anterior to isocenter.[7]
| Results and Discussion | | |
Planning results
Dose statistics of treatment plans created in TPS for multitarget, prostate, head and neck, and C shape are given in [Table 1]. Dose achieved in planning at our institute is closer to the planning goal as compared to planning results in the AAPM report. An individual can have his/her own way of making a plan on TPS, and hence, dose achieved could differ for different planner or TPS. While creating test plans in planning system, plan goal of all plans has been achieved, except in case of C shape harder where we have to keep core D10 <1000; similar planning result has been reported in the AAPM TG119 report. Our planning results are comparable with TG119 results which prove that the degree of beam modulation is similar. All test plans have been created on scan of slab phantom which is cuboidal and homogeneous. However, actual patient shape is irregular and inhomogeneity is certain. Hence, the current method validates the TPS commissioning accuracy but fails to facilitate actual treatment condition
Ion chamber measurements
Results of ion chamber measurements for high-dose points in PTV and low-dose points in avoidance structure are summarized in [Table 2] and [Table 3]. | Table 2: High dose points in the planning target volume measured with ion chamber
Click here to view |
 | Table 3: Low dose points in the avoidance structure measured with ion chamber
Click here to view |
Dose variation and CL have been calculated using following formula:
Dose variation = (measured dose − planned dose)/prescribed dose.[7]
CL = ǀmeanǀ + 1.96σ.[15],[16]
An analysis of this confidence limit has been done by Knill and Snyder.[17]
All measurements were repeated for 5–6 times for reproducibility of results. Our results for target dose and low dose avoidance structures measurement are better than the TG119 report. AAPM report's data are more scattered and therefore having greater standard deviation (SD). This is because AAPM TG119 report is a multiple institute planning report. On the other hand, our data are less scattered and hence lead to lesser SD. Although our study is a single institute, this shows reproducibility of our results. A variation in results may be seen if a different planner do separate planning and measurements on the same system as observed by Gordon et al.[18] However, all measurements were taken at planned gantry and collimator angles, but still phantom is a stationary while there are many motions in actual patient. These motions may cause a significant uncertainty which will further increase mean variation/SD and hence CL. In test plans, we have measured dose for a very smaller volume and fields are comparatively small as compared to actual clinical plans, in which fields are very large, especially in case of the head and neck. For carcinoma of the cervix and endometrium case where nodes are involved, the area to be irradiated is very large and hence results may differ a lot, i.e. current data could be different if we measure dose for large volume. In the IMRT process, which is basically complex field with nonuniform fluence, ideally planar and point dose measurements in region of high dose and low gradient are recommended.[19] Palta et al.[16] have given recommendation for ion chamber point dose measurement action level. They suggested the limit of 5% deviation in high dose-low gradient region and 7% in low dose-low gradient region. The European Society for Therapeutic Radiation Oncology booklet on “Guidelines for the Verification of IMRT”[20] summarizes the practice of several European institutions and also discusses the use of CL as expressed in TG119. They recommend tolerance limits of ± 3% for ion chamber measurement in target areas and action limits of ± 5% for point dose verification. AAPM TG119 has reported 4.5% CL in high dose region i.e. target structure and 4.7% in low dose region i.e. avoidance structure for ion chamber measurements. Our results for point dose measurements are much better agreement with acceptable limit.
Per field measurement
Per field measurement was done using portal dosimetry with gamma passing criteria of 3% dose difference and 3 mm DTA. Perusal of data is presented in [Table 4]; it is observed that our individual as well as overall gamma passing is well confined in CL recommended by the AAPM TG119 report. All the test plans have minimum gamma passing in > 98%, showing a good agreement between calculated and measured fluence map. As the AAPM report is a multiple institute study, where different institutes have used different modalities, a few have used diode array for assessing per field quality; therefore, CL is wider in the report. Commercially available diode- or ion chamber-based array has poor spatial resolution than film and EPID. Therefore, QA using these devices resulted in wider overall CL in TG119 report. We have used EPID for gamma evaluation, and EPID has greater spatial resolution than an array of diodes or ion chambers; hence, our percentage gamma passing is much larger than those from the ion chamber or diode array. If an institute uses EPID for gamma evaluation, then they may compare their results with our results of 99.5% (SD 0.33) mean gamma passing with CL 1.13. Gamma evaluation is now a standard method of testing prior to accepting a dose distribution.[21] Depuydt et al.[22] used gamma evaluation algorithm to compare efficiently calculated versus measured IMRT dose distribution. A comparative study of gamma index analysis for different commercial IMRT QA systems has been carried out by Hussein et al.[23]They have reported that for the same gamma passing criteria, different devices and software combinations resulted in different values of agreement with the measured gamma analysis. Since TG119 does not provide any criteria of gamma evaluation and CL for EPID, hence we have compared our data with the institute which have used EPID in this report and our data is consistent with that institute result.
To see the effect of measurements done at planned gantry and collimator angle on phantom, we compared those results with zero gantry and collimator angle. It was similar to one done for IMRT QA by measuring point dose variation and portal dose calculation.[24] For that, ten IMRT actual patient's plans were selected at random and average variation in dose and gamma passing has been reported. As a routine IMRT plans QA, this measurement has been carried out at zero gantry and collimator angle. The result is summarized in [Table 5]. | Table 5: Point dose variation and portal dose calculation for ten random patients
Click here to view |
If we compare these patient-specific IMRT QA results with the AAPM phantom study results, we see that there is insignificant difference between the two. Results of the study with planned gantry and collimator angle done on phantom are closer to their expected value as compared with the zero gantry and collimator angle done for patient-specific QA. The reason for the better results on phantom cannot be explained with confidence. One of the reasons could be due to easier planning on phantom structure in a homogenous medium, or other could be the smaller volume of targets in test plans as compared to contours in actual patient. A heterogeneous phantom can be used to evaluate the effect of density variation as is used by Gurjar et al.[25] The sources of deviation between planned and measured doses arise majorly due to inappropriate TPS commissioning and then because of delivery system and measurement process. The AAPM TG119 test suit used to validate the IMRT commissioning in this study can assess overall uncertainty in the system but cannot distinguish the source of error for that uncertainty. For a comprehensive analysis, this validation process may be clubbed with more IMRT QA tests and also wider test suit may be generated which resembles closely to clinical cases.
| Conclusions | | |
The validation of TPS commissioning with the AAPM TG 119 report will ensure accurate dosimetry and self-reliance on planning and measurement. Our overall results for planning and measurements are better than AAPM TG119 report. Portal dosimetry results have made CL narrower for EPID users. We used specifically EPID for per field measurement while the AAPM TG119 report had not derived CL for EPID separately, so this paper will provide a separate CL for gamma evaluation by EPID. Test suit gives a simple, basic, and initial testing tool for inter-comparison of individual results after TPS commissioning with the international one. More complicated broader and anatomical sight-specific test plans can be created and compared among different institutions for validation and improvement in commissioning so that patient can be benefitted by correct dosimetry.
Acknowledgments
We thank Dr. S. Hukku and Dr. S. Halder, Department of Radiation Oncology, Dr. B. L. Kapur Memorial Hospital, New Delhi, for their encouragement and support to work effectively. We also thank Roentgen Oncologic Solutions Pvt. Ltd., C/o Dr. B. L. Kapur Memorial Hospital, New Delhi, for providing support of equipment to carryout research work. We thank unknown reviewers who have helped us to improve the manuscript.
Financial support and sponsorship
Equipment support was provided by Roentgen Oncologic Solutions Pvt. Ltd., C/o Dr. B. L. Kapur Memorial Hospital, New Delhi.
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | Vergeer MR, Doornaert PA, Rietveld DH, Leemans CR, Slotman BJ, Langendijk JA. Intensity-modulated radiotherapy reduces radiation-induced morbidity and improves health-related quality of life: Results of a nonrandomized prospective study using a standardized follow-up program. Int J Radiat Oncol Biol Phys 2009;74:1-8.  |
| 2. | Kam MK, Chau RM, Suen J, Choi PH, Teo PM. Intensity-modulated radiotherapy in nasopharyngeal carcinoma: Dosimetric advantage over conventional plans and feasibility of dose escalation. Int J Radiat Oncol Biol Phys 2003;56:145-57. |
| 3. | Portelance L, Chao KS, Grigsby PW, Bennet H, Low D. Intensity-modulated radiation therapy (IMRT) reduces small bowel, rectum, and bladder doses in patients with cervical cancer receiving pelvic and para-aortic irradiation. Int J Radiat Oncol Biol Phys 2001;51:261-6. |
| 4. | Cozzi L, Fogliata A, Bolsi A, Nicolini G, Bernier J. Three-dimensional conformal vs. intensity-modulated radiotherapy in head-and-neck cancer patients: Comparative analysis of dosimetric and technical parameters. Int J Radiat Oncol Biol Phys 2004;58:617-24. |
| 5. | Laskar S, Bahl G, Muckaden M, Pai SK, Gupta T, Banavali S, et al. Nasopharyngeal carcinoma in children: Comparison of conventional and intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 2008;72:728-36. |
| 6. | Xing L, Curran B, Hill R, Holmes T, Ma L, Forster KM, et al. Dosimetric verification of a commercial inverse treatment planning system. Phys Med Biol 1999;44:463-78. |
| 7. | Ezzell GA, Burmeister JW, Dogan N, LoSasso TJ, Mechalakos JG, Mihailidis D, et al. IMRT commissioning: Multiple institution planning and dosimetry comparisons, a report from AAPM Task Group 119. Med Phys 2009;36:5359-73. |
| 8. | Ezzell GA, Galvin JM, Low D, Palta JR, Rosen I, Sharpe MB, et al. Guidance document on delivery, treatment planning, and clinical implementation of IMRT: Report of the IMRT Subcommittee of the AAPM Radiation Therapy Committee. Med Phys 2003;30:2089-115. |
| 9. | LoSasso T, Chui CS, Ling CC. Comprehensive quality assurance for the delivery of intensity modulated radiotherapy with a multileaf collimator used in the dynamic mode. Med Phys 2001;28:2209-19. |
| 10. | Das IJ, Cheng CW, Chopra KL, Mitra RK, Srivastava SP, Glatstein E. Intensity-modulated radiation therapy dose prescription, recording, and delivery: Patterns of variability among institutions and treatment planning systems. J Natl Cancer Inst 2008;100:300-7. |
| 11. | Ibbott GS, Followill DS, Molineu HA, Lowenstein JR, Alvarez PE, Roll JE. Challenges in credentialing institutions and participants in advanced technology multi-institutional clinical trials. Int J Radiat Oncol Biol Phys 2008;71 1 Suppl: S71-5. |
| 12. | Low DA, Harms WB, Mutic S, Purdy JA. A technique for the quantitative evaluation of dose distributions. Med Phys 1998;25:656-61. |
| 13. | Van Esch A, Tillikainen L, Pyykkonen J, Tenhunen M, Helminen H, Siljamäki S, et al. Testing of the analytical anisotropic algorithm for photon dose calculation. Med Phys 2006;33:4130-48. |
| 14. | Breitman K, Rathee S, Newcomb C, Murray B, Robinson D, Field C, et al. Experimental validation of the Eclipse AAA algorithm. J Appl Clin Med Phys 2007;8:76-92. |
| 15. | Venselaar J, Welleweerd H, Mijnheer B. Tolerances for the accuracy of photon beam dose calculations of treatment planning systems. Radiother Oncol 2001;60:191-201. |
| 16. | Palta JR, Kim S, Li JG, Liu C. Tolerance limits and action levels for planning and delivery of IMRT. In: Intensity-Modulated Radiation Therapy: The State Of The Art. American Association of Physicists in Medicine Medical Physics Monograph No. 29, Medical Physics Publishing, Madison, WI, USA. 2003. p. 593-612. |
| 17. | Knill C, Snyder M. An analysis of confidence limit calculations used in AAPM Task Group No. 119. Med Phys 2011;38:1779-84. |
| 18. | Gordon JD, Krafft SP, Jang S, Smith-Raymond L, Stevie MY, Hamilton RJ. Confidence limit variation for a single IMRT system following the TG119 protocol. Med Phys 2011;38:1641-8. |
| 19. | Fraass B, Doppke K, Hunt M, Kutcher G, Starkschall G, Stern R, et al. American Association of Physicists in Medicine Radiation Therapy Committee Task Group 53: Quality assurance for clinical radiotherapy treatment planning. Med Phys 1998;25:1773-829. |
| 20. | Alber M, Broggi S, Wagter CD, Eichwurzel I, Engstrom P, Fiorino C, et al. Guidelines for the verification of IMRT. Brussels, Belgium: ESTRO; 2008. |
| 21. | Janusz W, Tomasz M, Karolina M, Barbara D. The gamma evaluation method as a routine QA procedure of IMRT. Rep Pract Oncol Radiother 2009;14:162-8. |
| 22. | Depuydt T, Van Esch A, Huyskens DP. A quantitative evaluation of IMRT dose distributions: Refinement and clinical assessment of the gamma evaluation. Radiother Oncol 2002;62:309-19. |
| 23. | Hussein M, Rowshanfarzad P, Ebert MA, Nisbet A, Clark CH. A comparison of the gamma index analysis in various commercial IMRT/VMAT QA systems. Radiother Oncol 2013;109:370-6. |
| 24. | Cilla S, Viola P, Azario L, Grimaldi L, Craus M, D'Onofrio G, et al. Comparison of measured and computed portal dose for IMRT treatment. J Appl Clin Med Phys 2006;7:65-79. |
| 25. | Gurjar OP, Mishra PK, Singh N, Bagdare P, Mishra SP. Dosimetric study for the development of heterogeneous chest phantom for the purpose of patient-specific quality assurance. Radiat Prot Environ 2015;38:139-43.  |
[Figure 1]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]
|
Search Pubmed for