Journal of Medical Physics
: 2013  |  Volume : 38  |  Issue : 2  |  Page : 67--73

A quantitative analysis of intensity-modulated radiation therapy plans and comparison of homogeneity indices for the treatment of gynecological cancers

Pushpraj Pathak, Sanjeev Vashisht 
 Department of Medical Physics, Jawaharlal Nehru Cancer Hospital and Research Centre, Bhopal, Madhya Pradesh, India

Correspondence Address:
Pushpraj Pathak
Jawaharlal Nehru Cancer Hospital and Research Centre, Idgah Hills, Bhopal, Madhya Pradesh


The aim of present study was to evaluate the intensity-modulated radiation therapy (IMRT) plans using different homogeneity and conformity indices in gynecological cancers, as well as to compare and find out the most reliable and accurate measure of the dose homogeneity among the available indices. In this study, a cohort of 12 patients were registered for evaluation, those receiving dynamic IMRT treatment on Clinac-2300C/D linear accelerator with 15-Mega Voltage (MV) photon beam. Dynamic IMRT plans were created on Eclipse treatment planning system with Helios dose volume optimization software. Homogeneity indices (HI) such as H index, modified H index, HI index, modified HI index, and S-index (sigma-index) proposed by M Yoon et al. (2007) were calculated and compared. The values of S-index vary from 1.63 to 2.99. The results indicate that the H and HI indices and their modified versions may not provide the correct dose homogeneity information, but the S-index provides accurate information about the dose homogeneity in the Planning Target Volume (PTV). Each plan was compared with 6-MV photon energy on the basis of S-index and conformity index (CI). Organs at risk (OAR) doses with 6-MV and 15-MV beams were also reported.

How to cite this article:
Pathak P, Vashisht S. A quantitative analysis of intensity-modulated radiation therapy plans and comparison of homogeneity indices for the treatment of gynecological cancers.J Med Phys 2013;38:67-73

How to cite this URL:
Pathak P, Vashisht S. A quantitative analysis of intensity-modulated radiation therapy plans and comparison of homogeneity indices for the treatment of gynecological cancers. J Med Phys [serial online] 2013 [cited 2021 Sep 24 ];38:67-73
Available from:

Full Text


The basic requirement of radiotherapy treatment is to deliver maximum dose to the target volume (tumor) and as low as possible to the surrounding normal tissue. This requirement could be easily fulfilled if the separation between tumor control probability (TCP) [1] and normal tissue complication probability (NTCP) plot would be wide enough, but unfortunately both curves overlap at higher doses. The aim of the new technique in radiotherapy is to widen the separation between the TCP and NTCP, and at the same time, uniform distribution of the dose throughout the tumor. Intensity-modulated radiation therapy (IMRT) [2] is one modality by which the reduction of NTCP and escalation of target dose is possible, but may produce less homogeneous dose distribution within the planning target volume (PTV) as compared to three-dimensional conformal radiation therapy (3D-CRT) and conventional technique. In modern radiation therapy, specification of absorbed dose to the relevant volumes rather than some points is critical to communicate the treatment intent. The reported absorbed dose should be descriptive of the absorbed dose in the volume. The radiobiological effects and dose homogeneity are interrelated. The concept of equivalent uniform dose (EUD) for tumors as proposed by Niemierko [3] is one of the methods used to show the relationship between the dose homogeneity and radiobiological effects. [4],[5],[6] The equivalent uniform dose is defined as the biologically equivalent dose that, if delivered uniformly, would lead to the same reduction in the tumor volume as the actual dose that has an inhomogeneous distribution. Therefore, if the homogeneity indices are properly calculated, then it may be closely related to the EUD. Yoon et al. [7] have evaluated the relationship between EUD and homogeneity index (S-index) for brain tumor and reported highly linear distribution. The dose-volume histogram (DVH) [8] is the key tool of the 3D-treatment planning system, which summarizes the entire plan into a single 2D graph, though unable to show the spatial distribution of doses. Hence, it is an excellent tool for evaluating as well as comparing the competing plans. Other than evaluation of the dose coverage, cumulative and differential DVH are used to quantify the degree of homogeneity in the plans. Several definitions of the homogeneity index have been proposed and used depending upon the radiotherapy modality. A perfectly homogeneous dose to the planning target volume (PTV) would be characterized by a spike in the differential DVH (dDVH) and the vertical drop of the cumulative DVH (cDVH) line at the prescribed absorbed dose. In practice, the dDVH for PTV in a treatment plan has near Gaussian shaped distribution around the mean absorbed dose. Measure of such a distribution is nothing but the standard deviation of the mean dose (D sd ), which directly reflects the degree of dose homogeneity in PTV, and this number is called the sigma-index (S-index). Therefore, as recommended by the International Commission on Radiation Units and Measurements (ICRU) Report 83, [9] in best circumstances, both mean absorbed dose to PTV and standard deviation of the mean dose would be reported. In the present study, we evaluated differentdose homogeneity indices [7],[8],[9],[10],[11] and conformity index [12],[13],[14],[15],[16],[17] for the IMRT plans using 15MV photon energy in cervical cancer and effectiveness of different proposed indices were explored. Each plan was compared with 6MV photon energy. S-index, conformity-index, and organs at risk (OAR) doses were also reported for both photon energies.

 Materials and Methods

In the present study, different-dose homogeneity indices were evaluated for 12 patients receiving dynamic IMRT treatment on Clinac 2300C/D linear accelerator (Varian Medical Systems, Palo Alto, USA) with 15MV photon energy. The IMRT plans were generated on Eclipse (Varian Medical Systems, USA) treatment planning system with Helios™ dose volume optimization (DVO) software and pencil beam convolution (PBC) algorithm for dose calculation. Each plan consisted of nine co-planar non-opposing fields with equally spaced gantry at 40-degree interval. Computed tomography (CT) scan data sets were used for the delineation of target (PTV) and OAR volumes. To compare the plan with 6MV photons, all planning parameters and constraints were kept same, except energy. The different indices used for evaluating the plans are defined as under:

H-index = D max /D p , where H-index is the conventionally used homogeneity index, D max is the maximum dose and D p is the prescribed dose.

mH-index = D 5 /D p is the modified homogeneity index. Where D 5 is the dose received by the 5% volume of PTV.

HI = (D 2 -D 98 )/D p is another homogeneity index proposed in ICRU-83, where D2 and D98 represent the doses received by 2% and 98% volumes of PTV, respectively.

mHI = (D 5 -D 95 )/D p is the modified HI, where D 5 and D 95 represent the doses received by the 5% and 95% volumes of PTV, respectively.

S-index (D sd ) = ∑SQRT[(D i -D mean ) 2 ] × v i /V, where D sd is the standard deviation of mean dose, D mean is the mean dose of PTV, v i is the ith volume element receiving at least D i dose, and V is the total volume. In contrast to the other homogeneity indices, S-index uses the dDVH curve instead of cDVH curve.

CI [12] is defined as ratio of volume of the body receiving the prescribed dose (Vp) to the volume of the PTV receiving the same dose (PTVp), i.e., CI = Vp/PTVp.

Doses received by the OARs as maximum dose, mean dose, and other dose levels according to the department protocol have been evaluated and tabulated for both 16MV and 6MV photon energies.


The planning data of 12 patients were analyzed and values of H-index, mH-index, HI-index, mHI-index, S-index, and CI were computed and tabulated. The H-index values vary from 1.05 (highly homogeneous) to 1.17 (less homogeneous). Values of H-index as shown in [Table 1] indicate that for patients 1 and 5, these were same, but the cDVH of patient 1 was better than that of patient 5, as shown in [Figure 1], and their corresponding differential DVH with S-index values [Figure 2]. There is about 19% variation between their corresponding S-index values as shown in [Table 1]. The mH-index values vary from 1.004 (highly homogeneous) to 1.113 (less homogeneous). The mH-index values of patients 1 and 7 were same, but their corresponding cDVH as shown in [Figure 3] and dDVH in [Figure 4] were not matching, because in calculation of mH-index, only the ratio of doses at two points on the cDVH curve as defined above is taken into consideration and not the entire curve. The S-index values show difference of about 20%. The values of HI-index vary from 0.065 (highly homogeneous) to 0.143 (less homogeneous). The values of HI-index of patients 3 and 8 are nearly same (0.092 and 0.091, respectively), but their corresponding cDVH values as shown in [Figure 5] and dDVH with S-index in [Figure 6] show much difference. The S-index values differ by 10%, which can be seen from [Table 1]. HI-index values, which are less than or close to 0.1 give better dose homogeneity to the PTV. The mHI-index values vary from 0.050 (highly homogeneous) to 0.097 (less homogeneous). The values of mHI-index of patients 1 and 4 are same, but their corresponding cDVH and dDVH values [Figure 7] and [Figure 8] show much variation, as shown in [Table 1].{Figure 1}{Figure 2}{Figure 3}{Figure 4}{Figure 5}{Figure 6}{Figure 7}{Figure 8}

S-index values show a difference of about 8%. The values of S-index vary from 1.63 to 2.99. Also, their corresponding conformity index (CI) values were tabulated, which vary from 1.0 (highly conformal) to 1.092 (less conformal). The value close to 1 gives better conformity of dose to PTV. All homogeneity indices and their modified versions mentioned above, except S-index, based on the limited point's ratio on the cDVH curve, may not give the correct information about the dose homogeneity. Hence, the S-index, which is based on the entire dDVH curve, gives the correct information about dose homogeneity in the target volume. It can be seen from [Table 1] that the values of homogeneity indices were improved by their modified version, but may not reflect the correct dose homogeneity. The S-index is nothing but the measure of the spread of the dDVH curve from their mean dose value, i.e, standard deviation of the mean dose. Also, from [Table 1], it can be seen that the S-index values obtained after analyzing the IMRT planning data of 12 patients in the present study is below 2.5, except for patients 5 and 7; for these patients, it is approximately 3. This shows that our plans fall under outstanding and good categories, except two plans, according to Yoon (2007) criteria. The values of S-index less than or close to 2 reflects better dose homogeneity in PTV. The S-index values of patients 2 and 6 are close to each other and the corresponding cDVH value as shown in [Figure 9] has a sharp fall, which reflects that whole target receiving nearly the same prescribed dose, i.e., homogeneous dose in the entire target volume. Further, it can be explained that if two cDVH curves appear same, their S-indices may not necessarily be close to each other, as shown in [Figure 1] and [Figure 2]. But, if two S-indices are close to each other, then their corresponding cDVH curves must be close to each other. From [Table 1] and [Table 2], it can be seen that S-index values are less in the 15MV plan in comparison to the 6MV plan, i.e., 15MV plans have more homogeneous dose distribution. In most cases, the OARs received lower dose in the 15MV plan in comparison to the 6MV plan, but were not clinically significant, as shown in [Table 3],[Table 4],[Table 5] and [Table 6] for bladder, rectum, femoral heads, and bowel, respectively. Also, it was observed that the integral dose outside the target volume is less in 15MV plans. Hence, two most important benefits of using 15MV beam for IMRT in the pelvic region include better homogeneous dose to the target volume and lower dose to the extra target volume. [Figure 10] shows the same level of homogeneity (close S-index values) for the 6MV and 15MV plans, but the peak position is at a slightly lower dose level than the prescribed one in case of the 6MV plan. It has been observed that for underweight patients, there is not much difference in S-index values in either the 6MV or 15MV IMRT plan for pelvic regions, but a high level of difference has been observed, especially in overweight patients.{Figure 9}{Figure 10}{Table 1}{Table 2}{Table 3}{Table 4}{Table 5}{Table 6}


The selection of dose reporting points lying within a high or low-absorbed dose-gradient region could then significantly misrepresent the absorbed dose throughout the volume concerned. A comparison between homogeneity indices indicate that the indices that are based on the limited points on the DVH curve may not be the actual measurement of the dose homogeneity. S-index may be the actual measure of the dose homogeneity. The modified homogeneity indices give some improved results, but still not matching with their cDVH curves and S-index for the two closed modified index values. Dose homogeneity in the patients at the time of dose delivery can only be assured if leaves are in the correct position at each moment regardless of the multi-leaf collimator (MLC) design and delivery methods. Because the dose delivered throughout the target volume with IMRT is sensitive to leaf positioning errors, 1 mm error in positioning will produce about 5% errors in dose delivery, as reported by Thomas LoSasso et al. [18] Hence, the aim is to maintain the positioning errors below 0.2 mm and to ensure this, proper quality assurance (QA) [18],[19],[20],[21] of the MLC must be carried out regularly; these may include picket fence and garden fence tests with and without intentionaly introduced errors. This test is able to detect error in positioning by 0.2 mm as we have observed. Further, Dynalog file viewer (Varian) is the software for position-verification and plays an important role in checking the positional error of the leaves. The minimum physical leaf gap is the closest separation maintained by the opposite leaf without wear and tear, which affects the overall dose delivery; for Varian MLC, this is about 0.5 mm. Dosimetric leaf gap is the most important parameter and attention must be paid to test and quantify the same, defined by the opposing leaf pairs of MLC for each beam. One of the methods used to evaluate the same is sweeping gap output for different slit widths formed by MLC and programmed in the MLC shaper software supplied by Varian Medical Systems, USA. The measured leaf gap values are 1.7 mm for 6 MV and 1.8 mm for 15 MV. Although using 15MV beam for IMRT is a controversial issue but has certain advantages and disadvantages. Although various advantages such as higher homogeneity and less integral dose have been discussed above, it has some disadvantages as it has higher probability to produce neutrons as compared to 10-MV beam, but less probability than 18-MV beam. Kry et al. [22] reported that contribution of neutrons in out-of-field dose equivalent is significant for beam >15 MV. In literature, [22],[23] it has been reported that the calculated risk of second malignancy is 3.4% using Varian 15MV beam for IMRT and 3.7% with Siemens 6MV beam. Hence, the contribution of neutrons is not significant with 15MV IMRT plans. The other concern is dose modulation and transmission. The leaf transmission in 15-MV Varian accelerator may not produce significant effect as total MU and time of delivery is less in comparison to 6-MV beam. The modulation in IMRT with higher beam is limited by lateral beam degradation due to production of high-energy electrons. However, high modulation and sharp fall in dose distribution is needed to avoid close and sensitive OARs especially those have less tolerance, whereas the pelvic region may not require the highest level of modulation. Also, it has been reported [24] that there is little difference in volume (closed to target) exposed to dose, regardless of energy and number of fields used, but significant increases in dose for distant volume occur in low energy/few fields (<9) plan as compared to high energy/many field (≥9) plan. Also, using high-energy beams in prostate IMRT is beneficial for saving some OARs even in low-dose regions under 50% of prescription dose, but losing benefits in high-dose regions may not be clinically significant. [25] Hence, the use of 15MV beam with ≥9 fields IMRT plan in case of cervical cancer especially in overweight patients are well justified.


In the present article, comparisons between the different homogeneity indices and their modified versions with their DVH and S-index values have been presented for both 6MV and 15MV beams. From the results, it can be seen that the S-index represents the correct dose homogeneity in the target. It can be concluded here that each IMRT plan must be evaluated and compared by using the S-index score because S-index is directly related to the biological effects (equivalent uniform dose).


1Baumann M, Petersen C. TCP and NTCP: A basic introduction. Rays 2005;30:99-104.
2Webb S. The physical basis of IMRT and inverse planning. Br J Radiol 2003;76:678-89.
3Niemierko A. Reporting and analyzing dose distributions: A concept of equivalent uniform dose. Med Phys 1997;24:103-10.
4Wu Q, Djajaputra D, Liu HH, Dong L, Mohan R, Wu Y. Dose sculpting with generalized equivalent uniform dose. Med Phys 2005;32:1387-96.
5Lee TF, Ting HM, Chao PJ, Wang HY, Shieh CS, Horng MF, et al. Dosimetric advantages of generalised equivalent uniform dose-based optimisation on dose-volume objectives in intensity-modulated radiotherapy planning for bilateral breast cancer. Br J Radiol 2012;85:1499-506.
6Henriquez FC, Castrillon SV. A quality index for equivalent uniform dose. J Med Phys 2011;36:126-32.
7Yoon M, Park SY, Shin D, Lee SB, Pyo HR, Kim DY, et al. A new homogeneity index based on statistical analysis of the dose-volume histrogram. J Appl Clin Med Phys 2007;8:9-17.
8TRS-430. Commissioning and quality assurance of computerized planning system for radiation treatment of cancer. Vienna: International Atomic Energy Agency (IAEA), Technical Report Series; 2004.
9ICRU Report 83. "Prescribing, recording, and reporting photon-beam intensity modulated radiotherapy (IMRT)". Oxford: Oxford University press, International Commission on Radiation Units and Measurements; 2010.
10Mock U, Georg D, Bogner J, Auberger T, Potter R. Treatment planning comparison of conventional, 3D conformal, and intensity modulated photons and proton therapy for paranasal sinus carcinoma. Int J Radiat Oncol Biol Phys 2004;58:147-54.
11Nutting CM, Rowbottom CG, Cosgrove VP, Henk JM, Dearnaley DP, Robinson MH, et al. Optimisation of radiotherapy for carcinoma of the parotid gland: A comparison of conventional, three-dimensional conformal, and intensity-modulated technique. Radiother Oncol 2001;60:163-72.
12ICRU Report 62 "Prescribing, recording, and reporting photon-beam therapy" (Supplement to ICRU Report 50). Bethesda, MD, USA: ICRU Publications; International Commission on Radiation Units and Measurements; 1999.
13Feuvtet L, Noël G, Mazeron JJ, Bey P. Conformity index: A review. Int J Radiat Oncol Biol Phys 2006;64:333-42.
14Pyakuryal A, Myint WK, Goppalkrishnan M, Jang S, Logemann JA, Mittal BB. A computational tool for the efficient analysis of dose-volume histogram for radiotherapy treatment plans. J Appl Clin Med Phys 2010;11:137-157.
15Lee EK, Fox T, Crocker I. Simultaneous beam geometry and intensity map optimization in intensity-modulated radiation therapy. Int J Radiat Oncol Biol Phys 2006;64:301-20.
16Kam 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.
17Oliver M, Ansbacher W, Beckham WA. Comparing planning time, delivery time and plan quality for IMRT, RapidArc and Tomotherapy. J Appl Clin Med Phys 2009;10:117-131.
18LoSasso T, Chui CS, Ling C. Physical and dosimetric aspects of a multileaf collimation system used in the dynamic mode for implementing intensity modulated radiotherapy. Med Phys 1998;25:1919-27.
19Sharma DS, Gupta T, Jalali R, Master Z, Phurailatpam RD, Sarin R. High-precision radiotherapy for craniospinal irradiation: Evaluation of three-dimensional conformal radiotherapy, intensity-modulated radiation therapy and helical tomotherapy. Br J Radiol 2009;82:1000-9.
20LoSasso T, Chui CS, Ling C. 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.
21Bayouth JE, Wendt D, Morrill SM. MLC quality assurance technique for IMRT applications. Med Phys 2003;30:743-50.
22Kry SF, Salehpour M, Followill DS, Stovall M, Kuban DA, White RA, et al. The calculated risk of fatal secondary malignancies from intensity-modulated radiation therapy. Int J Radiat Oncol Biol Phys 2005;62:1195-203.
23Welsh JS, Mackie TR, Limmer JP. High-energy photons in IMRT: Uncertainities and risks for questionable gain. Technol Cancer Res Treat 2007;6:147-9.
24Pirzkall A, Carol MP, Pickett B, Xia P, Roach M 3 rd , Verhey LJ. The effect of beam energy and number of fields on photon-based IMRT for deep-seated targets. Int J Radiat Oncol Biol Phys 2002;52:434-42.
25Park JM, Choi CH, Ha SW, Ye SJ. The dosimetric effect of mixed-energy IMRT plans for prostate cancer. J Appl Clin Med Phys 2011;12:147-157.