Skip to main content
Journal of Medical Physics logoLink to Journal of Medical Physics
. 2018 Nov;43(Suppl 1):S14–S38.

Oral

PMCID: PMC6225529
J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-1: A RADIOBIOLOGICAL AND DOSIMETRIC COMPARISON BETWEEN SIMULTANEOUS INTEGRATED AND SEQUENTIAL BOOST INTENSITY MODULATED ARC TREATMENT OF LOCALLY ADVANCED HEAD AND NECK CANCER: A PART OF RANDOMIZED PROSPECTIVE CLINICAL STUDY

Abhijit Mandal 1, Sunil Choudhary 1, Nilesh Mani 1

Introduction: Intensity Modulated Arc Therapy (IMAT/VMAT/Rapid Arc) is now considered as a better choice of intensity modulated radiotherapy treatment technique because of its advantages like higher patient throughput, expenses of lower monitor units and simplicity in treatment execution. The treatment delivery regimen may be in the form of Sequential Boost (SEQB) or Simultaneous Integrated Boost (SIB). In the SEQB regimen different plans are used for each phase of treatment, where as in SIB regimen a single plan is used throughout the course of treatment delivering differential doses to different target volumes.

Purpose: To compare the radiobiological and dosimetric parameters of the two regimens of IMAT i.e. SIB versus SEQB in locally advanced head and neck cancer patients.

Materials and Methods: A total of 24 previously untreated locally advanced head and neck cancer patients were included in this study. The patients were prospectively randomized into SIB and SEQB arm. The CT data set in treatment position were transferred to the ECLIPSETM Treatment Planning System (Version 11.0.47). All the target volumes i.e. Gross Target Volume (GTV), High Risk (HR), Intermediate Risk (IR), Low Risk (LR) Planning Target Volume (PTV) and Organ at Risk (OAR) were delineated. The treatment plan of each phase in SEQB regimen were independently optimized. All the patients were treated using 6MV linear accelerator (UNIQUETM Performance, Varian Medical System).

Results: In the SIB arm 11 patients and in the SEQB arm 13 patients were enrolled. The BED (10) value for HR PTV was same in both group, whereas for IR PTV and LR PTV the values were 59.0, 63.6 and 50.0, 56.0 for SIB and SEQB arm respectively. The V (95) values were 100% for all the target volumes in both arms of patents. The average D (100) value for GTV, HR PTV an IR PTV were higher in SEQB arm than in SIB arm (7066cGy vs. 6900cGy, 6720cGy vs. 6497cGy and 6308cGy vs. 5917cGy). The average D (100) value for LR PTV were 5037cGy, 4871cGy for SIB and SEQB arm respectively. The spinal cord maximum doses were within the tolerance in both group of patients. The left and right parotid gland sparing was comparable in both group of patients. Average integral dose was 12.8% higher in SIB group than SEQB group. The average total monitor unit per fraction was 25.6% higher in SEQB arm than in SIB arm.

Discussion: The SIB treatment regimen offers more organizational benefits over SEQB regimen in terms of practicality and lesser chances of treatment uncertainty. But SIB treatment when combined with chemotherapy may increase the toxicity profile as larger volume is irradiated throughout the treatment course. To balance the radiobiological volume effect, BED (10) values for IR PTV and LR PTV intentionally kept lower in SIB arm. In the SEQB, each phases were optimized independently, which cause the higher D (100) values for GTV, HR PTV and IR PTV in the plan sum dose volume histogram. D (100) value of LR PTV is lower in SEQB arm as prescription dose is much lesser than SIB arm (5000cGy vs. 5400cGy). As large volume is irradiated throughout treatment course in SIB arm, a clear increase in integral dose is observed. Promising reduction in MUs in SIB treatment may be considered as a good merit of this regimen.

Conclusion: SIB treatment regimen may be considered as more logical and efficient in the treatment of locally advanced head and neck caner with comparable radiobiological and dosimetric parameters. The clinical comparison may explore more pros and cons of SIB and SEQB.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-2: MEASUREMENT AND ANALYSIS OF SURFACE DOSE FOR FLATTENING FILTER FREE MEDICAL ACCELERATOR USING ADVANCED MARKUS CHAMBER

Bibekananda Mishra 1,4, A Pichandi 2, S D Sharma 3,4, T Palani Selvam 3,4

Introduction: Surface dose plays a significant role in radiotherapy. Doses received by the basal skin layer can result in complications such as skin erythema, epilation, dry desquamation, wet desquamation, necrosis etc depending on the magnitude of doses received.[1] Hence, accurate measurement of dose at the surface is essential for proper treatment of patients. Surface dose is machine dependent and can be affected by many parameters such as field size, source-to-surface distance (SSD), beam energy, materials present in the beam line, type of dosimeter used for its measurement etc.[2]

Purpose: Recently, flattening filter-free (FFF) medical linear accelerator (linac) has been introduced in radiotherapy. With the filter removed, this low-energy component is allowed to pass through to the patient and will act to increase the surface dose. Energy spectrum and electron contamination are the two factors, which can change the surface dose in FFF. In this study, the relative surface dose has been studied for two different linac models in FF and FFF mode for 6 MV beam energy considering the fact that 6 MV is the beam of choice for various clinical cases.

Materials and Methods: Measurements are carried out for photon beams of one nominal energ 6 MV generated from Elekta Ltd make Versa HD model and Varian Medical System make TrueBeam Linac model in FF and FFF mode. The plane parallel plate ionization chamber, PTW Germany make Advanced Markus chamber Ionization chamber type 34045, having volume 0.02 cm3 with a thin flat window thickness of 0.03 mm (membrane material: PE (CH2)), a window area density of 2.76 mg/cm2, a plate separation of 1 mm and a collecting electrode material acrylic (PMMA) graphite-coated having diameter of 5 mm was used. The surface dose for any field size is defined as the dose measured at 0.5 cm depth (from the surface) for that field size divided by the dose at dmax at a 10×10 cm2 field size [3]. Surface doses are expressed relative to the dose at dose maximum for the respective energy and field size. The measurements were done for field sizes of 5x5 cm2, 10x10 cm2, 15x15 cm2, 20x20 cm2 and 25x25 cm2 with build-up depths extending from the surface to 2 mm towards the dmax and at dmax. As the window thickness of the chamber is very less in comparison to other chambers the positional accuracy is considerable. The PP chamber was placed at 0mm, 1 mm and 2 mm depth to interpolate the ionization reading at 0.5 mm depth. The ionization value was also recorded for dmax of respective field sizes at 100 cm Source to Surface Distance (SSD).

Results and Discussion: The relative surface dose was observed to be greater for the FFF beam as compared to the flattened beam for the photon beam energy 6 MV in case of Varian True beam Linac. However, for Elekta Versa HD, the trend was similar up to 15x15 cm2 field size and after which the relative surface dose increases as compared to 6FFF. TrueBeam gives higher surface dose than Versa HD for all the field sizes. Table 1 presents the measured relative surface doses from the two accelerators.

Table 1.

Measured relative surface dose (%) for 6 MV x-rays of Versa HD and TrueBeam medical accelerators

graphic file with name JMP-43-14-g001.jpg

Conclusions: FFF beam change the dosimetric characteristics of photon beams by softening the energy spectra compared to beam hardened FF beams, thus changing surface doses. Our study compares the surface dose for two different Linac model from two different manufacturer. It is found that the variation in surface dose in Versa HD Linac is less in comparison to TrueBeam Linac.

References

  • 1.Carl J, Vestergaard A. Skin damage probabilities using fixation materials in high-energy photon beams. Radiothe. Oncol. 2000;55:191–8. doi: 10.1016/s0167-8140(00)00177-8. [DOI] [PubMed] [Google Scholar]
  • 2.Kry SF, Smith SA, Weathers R, Stovall M. Skin dose during radiotherapy: a summary and general estimation technique. J Appl. Clin. Med. Phys. 2012;13(3):20–34. doi: 10.1120/jacmp.v13i3.3734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.International Electrotechnical Commission (IEC) 60601-2-1, Geneva, Switzerland. 2010 [Google Scholar]
J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-3: A METHOD TO PREDICT ACHIEVABILITY OF CLINICAL OBJECTIVES IN IMRT

R Vaitheeswaran 1, Bojarajan Perumal 1, K J Maria Das 1, S Harikrishna Etti 1, Natarajan Ramar 1, S R Meher 1

Purpose: A data-driven method to predict the achievability of clinical objectives before performing the IMRT optimization is proposed.

Materials and Methods: In our approach, “Geometric Complexity (GC)” is computed to estimate the achievability of clinical objectives. Here, GC is the measure of the number of “unmodulated” beamlets or rays that intersect the Region-of-interest (ROI) and the target volume. The geometric complexity ratio (GCratio) between the GC of a region of interest (ROI) in a reference plan and the GC of the same ROI in a given plan is computed. The GCratio of a ROI indicates the relative geometric complexity of the ROI as compared to the same ROI in the reference plan. Hence, by using GCratio it is possible to predict the achievability of clinical objective associated with the ROI optimizer. Basically the likelihood for the optimizer to achieve the clinical objective defined for a given ROI is inversely proportional to GCratio. We have evaluated the proposed algorithm on six Head and Neck cases using Pinnacle3 (version 9.10.0) TPS.

Results and Discussion: Out of the total of 42 clinical objectives from six cases accounted in the study, 37 were in agreement with the prediction, which implies an agreement of about 88% between predicted and obtained results. The Pearson correlation test shows a positive correlation between predicted and obtained results (Correlation = 0.81, r2 = 0.66, P < 0.005).

Conclusion: The study demonstrates the feasibility of the proposed method in head and neck cases for predicting the achievability of clinical objectives with reasonable accuracy.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-4: DOSIMETRIC VERIFICATION OF 3DCRT/VMAT USING INDIGENOUSLY DEVELOPED CU DOPED LITHIUM TETRABORATE (LI2B4O7:CU) THERMOLUMINISCENT CRYSTALS

Ankit Srivastava 1, Rahul Kumar Chaudhary 1, Babita Tiwari 2, Rajesh Kumar 1, Vikram Mittal 3, Sudesh Deshpande 3, S D Sharma 1,4, SC Gadkari 2, D Datta 1,4

Introduction: Dose verification in the radiotherapy plays a very important role in overall success of different radiotherapy techniques. Since it is necessary to do dose verification for advanced treatment techniques where verification of planned MUs is not possible with manual calculations, hence it is followed regularly at most of the radiotherapy centre. Different type of dosimetry system are used for this purpose, however, thermoluminescent dosimeters (TLDs) features many advantage such as small detector size, close tissue equivalence, high sensitivity etc. and hence, still preferred over other dosimetry system for dose verification in radiotherapy [1]. The LiF:Mg, Ti (TLD-100) is the most chosen dosimeter in radiotherapy. Technical Physics Division of Bhabha Atomic Research Centre has developed a high sensitive, tissue equivalent Li2B4O7:Cu TL crystal for dosimetric applications [2]. The aim of this work was to study the potential application of the indigenously developed Li2B4O7:Cu TL dosimeters in dosimetric verifications of different treatment techniques used in external beam radiotherapy.

Materials and Methods: Technical Physics Division of Bhabha Atomic Research Centre has developed TLDs that are available in form of chip of dimension 3 x 3 x 1 mm3. TLDs used in the study were selected on the basis of repeatability test. Repeatability of the TLDs was checked out by irradiating freshly annealed TLD chips to a dose of 200 cGy using 6 MV X-ray. The process was repeated 4 times. The mean TL read out value, its standard deviation and coefficient of variation were calculated for each TLD. The TLD chips with coefficient of variation under 1.5% were selected for further study. The selected TLDs were subjected to further tests such as linearity with dose, energy dependence, dose rate dependence, dose fractionation dependence, angular dependence etc. The calibrations of TLDs were done by calculating elemental correction coefficient for each TLD. Finally TLDs were used for patient specific dosimetric QA for different planning techniques. The TLD-100 samples were, in parallel, also evaluated for above mentioned tests. Indigenously developed anthropomorphic IMRT dosimetry QA phantom representing the thorax region was used. The phantom has in-built planning target volume (PTV) as well as organ at risk (OAR) with feature to carry out point as well as 2-D dosimetric QA using TLD/Optical stimulated luminescence dosimeter (OSLD) and radiochromic films respectively. Volumetric CT scan of the phantom was acquired with slice thickness of 2.5 mm at 120 kVp using helical CT machine. Dose Verification QA plans for different techniques (3DCRT and VMAT) were made using Varian's Eclipse treatment planning system. The phantom containing TLDs in-situ at different positions inside the PTV as well as OAR regions were irradiated for the planned treatment procedure. The TLD measured dose at all positions in the PTV as well as OAR were compared to TPS calculated dose at corresponding locations and relative percentage deviations with respect to prescribed dose was determined.

Results: The percentage variations in 3DCRT were found to be within 3%, both in PTV and OARs. However, percentage variations in VMAT were found to be within 3% at most of the points except at one point in OAR, where it was found to be 5.76%. This higher variation may be attributed to the higher dose gradient in OAR.

Conclusions: The results of dose verification, carried out using Li2B4O7:Cu in 3DCRT/VMAT were found to be satisfactory. These initial enthusiastic results also prompt to carry out further detailed study on the potential application of these crystals as a substitute of TLD-100 in much wider spectrum of medical dosimetry.

References

  • 1.Kron T. Applications of Thermoluminescence Dosimetry in Medicine. Radiat. Prot. Dosim. 1999;85(1-4):333–340. [Google Scholar]
  • 2.Tiwari B, Rawat NS, Desai DG, Singh SG, Tyagi M, Ratna P, et al. Thermoluminescence studies on Cu-doped Li2B4O7 single crystals. J Lumin. 2010;130(11):2076–2083. [Google Scholar]
J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-5: IMPACT OF ARTIFACT EFFECT IN THE PLANNING OF VMAT AND HELICAL TOMOTHERAPY PLAN ANALYSES IN THE HEAD AND NECK CASES AND ITS COMPARISON

T Saminathan 1, P Velliangiri 1

Introduction: The metal implant in the planning Computerized Tomography (CT) image is unable to be avoidable. The presence of metal implant leads to artifact in the planning CT image due to its Extreme high absorption, Artifact effect leads to loss of anatomical information and inaccuracy in the dose calculation.

Purpose: The aim of the study is to analysis the impact of artifact effect in the planning of VMAT and Helical Tomotherapy (HT) planning in the Head & Neck (HN) cases and to compare both VMAT and HT plans.

Materials and Methods: There are six patients have under gone planning CT image with and without manual inserted metal (Unknown density) from 15 total patients, and remaining 7 patients image data were selected with unavoidable metal like dental implant and fibula graft. The First seven patient's planning have done with both machine True Beam (TB) Linac (Varian Medical system) equipped with Millennium 120 MLC and Tomo (H) (Tomotherapy-accuray) having 64 binary MLCs, Each patient's plans have done with three different categories are With Artifact (W-ART), Without Artifact (WO-ART) and HU corrected (WC-ART), The second Each 7 patients plans done with two categories W-ART and WC-ART. Eclipse planning system (V13.6) and Tomotherapy planning system were used for All planning for both TB and Tomo (H) respectively. We were measured dosimetry data for all plans for comparison between Eclipse plan and Helical Tomotherapy (HT) plan.

Results: We were found significant variation in Conformity Index (CI) and homogeneity index (HI) between W-ART and WO-ART in TB and HT, and less than 1.5% variation between W-ART and WC-ART in both machine. Insignificant variation was found in OAR doses between W-ART and WC-ART. Almost 3-4% more deviation was found in OAR doses between W-ART and WO-ART in both machines. HT plans having better 98% dose coverage than TB patients plans. Avg 2-3Gy dose deviation was found more in TB than HT plans.

Discussion and Conclusion: True beam plans are better CI and HI than HT plans in image having with artifact. HT plans having better dose coverage and HI than true beam in image without artifact. HT plans are better dose reduction in OAR Dose than TB planning for both images having with and without artifact. There is no significant variation was found between corrected HU plans and W-ART plans.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-6: EVALUATION OF VOLUMETRIC MODULATED ARC THERAPY FOR THE CARCINOMA OF LUNG

S Sriram Prasath 1, Arun Balakrishnan 1, A Melinda Monica 1, M Vidhya Shree 1, Shakambhari Sadangi 1, Soumen Guha 1, S E Koushik 1

Introduction: In patients with locally advanced non metastatic carcinoma of lung with large Planning Target Volume (PTV) eligible for radical radiotherapy, meeting dose constraints for organ at risks (OAR) with Three Dimensional Conformal Radiotherapy Therapy (3DCRT) is not possible and in these patients VMAT is necessary to achieve this.

Purpose: The main goal of the study is to analyse the dosimetric parameters of radiotherapy planning with reference to International Commission on Radiation Units and Measurements (ICRU 83) for carcinoma of lung treated using Volumetric Modulated Arc Therapy (VMAT) on Varian Eclipse Treatment Planning Systems (TPS) version 15.1.

Materials and Methods: Thirty patients of Lung Cancer with Median Volume of (PTV) 784.71 cc were planned using the technique VMAT, in the Eclipse version 15.1 Anisotropic Analytical Algorithm (AAA). These plans were evaluated using dosimetric parameters Conformity Index (CI), Homogeneity Index (HI) as recommended in (ICRU 83).

Results: The median CI and HI for the VMAT plans were 0.89, 0.11 respectively. The Median Mean lung doses (MLD) was 15.84 Gy. Lung Median V20Gy, V10Gy, V5Gy were 29.88%, 46.6%, 64.74%. Heart Median V30Gy was 25.54%. Heart Median Mean doses was 18.13 Gy. Spinal Canal PRV Median Max Doses were 43.84 Gy. Esophagus Median Max dose were 60.36 Gy. Esophagus Median V15Gy was 52.76%.

Discussion and Conclusion: Even though We could achieve the good values in dosimetric parameters like CI, HI for the PTV Coverage normal tissues constraints for the lesser PTV volumes in the Field-in-Field Plans, VMAT plans seems to be better when the PTV volumes were larger and if were very much close to the Spinal Cord and concavity nature. VMAT enables delivery of radical doses of radiotherapy even in patients with very large PTVs where dose constraints could not be met with 3DCRT.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-7: OPTICALLY STIMULATED LUMINESCENCE OF LICAALF6: TB PHOSPHOR

Pooja Seth 1, Anuj Soni 1, Shruti Aggarwal 1

Introduction: The measurement of the radiation dose using highly sensitive TL/OSL radiation dosimeters with great precision and accuracy is increasing worldwide. In medical dosimetry, it is important to precisely measure and monitor the dose delivered to the patients to ensure that prescribed dose is received by the target area and normal tissue is within tolerance. OSL has become a well known technique to measure the radiation dose absorbed by the material. LiCaAlF6 is one of the important materials used as OSL phosphor due to its interesting properties such as non toxicity, low hygroscopicity, high thermal durability, wide band gap etc. LiCaAlF6:Eu2+ and LiCaAlF6:Ce3+ have shown good TL and OSL properties. Therefore, it is important to improve the OSL dosimetric properties of this phosphor doping with other rare earth ion dopant. In this work, we have attempted to prepare the LiCaAlF6 phosphor doped with Tb and investigate its OSL properties.

Objectives: To prepare the LiCaAlF6 phosphor doped with Tb using melting method and investigates the optically stimulated luminescence (OSL) properties.

Materials and Methods: LiCaAlF6: Tb was prepared by using reagent grade salts of the constituent metals and rare earth salt Tb in the form of TbF3 (Terbium Fluoride) in the stoichiometric ratios in a graphite crucible and heated in the furnace available in SERL lab at GGSIPU at ~ 1073K in argon gas atmosphere. The samples were annealed at 673K for 1 h before use. The CW-OSL was recorded on Risø TL/OSL reader after stimulated with blue light with a dose of 20 mGy.

Results and Discussion: The XRD pattern matches well with the standard XRD pattern (pdf no. 00-043-1481) which shows the formation of pure LiCaAlF6 phase. Optically stimulated luminescence - The OSL signal of LiCaAlF6 and LiCaAlF6: Tb prepared as described above compared with the OSL signal from Al2O3: C. The OSL intensity of LiCaAlF6: Tb is 25 times higher than undoped LiCaAlF6. It is worth mentioning that OSL intensity of LiCaAlF6: Tb was also higher about 1.5 times as compared with commercially available Al2O3: C (BARC), under the same measurement conditions explained earlier.

Conclusion: LiCaAlF6 and LiCaAlF6: Tb phosphor were synthesised by melting method and their OSL properties were investigated. The OSL intensity of LiCaAlF6: Tb is 25 times higher than undoped LiCaAlF6. It is worth mentioning that OSL intensity of LiCaAlF6: Tb was also higher about 1.5 times as compared with commercially available Al2O3: C (BARC). These promising results led to further investigation of the material and applicability of this phosphor in medical dosimetry.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-8: OUT OF FIELD DOSE MEASUREMENTS FOR ELECTRON BEAMS AT DIFFERENT DEPTHS AND DISTANCES

M Ragu 1, K Midun 1

Introduction: The large improvements in radiotherapy (RT) procedure have led to high survival rates of the patients, so the possible late side effects of the dose delivered to the normal tissues will be a growing concern. More understanding of side effects of RT will require not only improved control of the high doses delivered to the target volumes, but also better knowledge of the unintended but unavoidable lower doses delivered out of the target. Due to scattering radiation, it affects the normal tissues and probable to induce the secondary cancer. In external beam therapy, out of field doses are one of the core contributors.

Purpose: To investigate the out of field dose for the electron beams used in external beam radiation therapy with different applicators size with different depths and distances.

Materials and Methods: For this experiment we are using Elekta synergy platform linear accelerator. It has three electron energies of 6 MeV, 12 MeV, and 15 MeV. These measurements were carried out with the parallel plate ionization chamber with standard imaging electrometer. The chamber was placed at isocenter in slab phantom with 2 cm backscatter. For all this experiment we delivered 100 MU. Readings are taken at various depths like 1 cm, 3 cm, 5 cm and 10 cm with different applicator sizes. The out of field dose measurements were taken by 5 cm to 50 cm distance from the field edge in cross plane and in-plane. All measurements are taken at 100 cm SSD.

Results: Out of field doses in high energy electron beams are evaluated for different depths (1 cm, 3 cm, 5 cm, and 10 cm) with different applicator sizes. The readings are taken at isocenter and various off- axis distance, 5 cm to 50 cm in cross plane (x) and in plane (y+, y-). The percentage of the dose evaluated by corresponding meter reading at isocenter. The peak value of out of field dose for 6 MeV, 12 MeV & 15 MeV tabulated for different applicator size at the following depths 1 cm, 3 cm, 5 cm and 10 cm and at 5 cm distance the peak value of out of field dose observed and tabulated.

Discussion and Conclusion: Graphs plotted between percentage dose with off axis distances. Percentage dose increases in cross plane (x-plane). For all three energies and all applicator sizes, out of field dose at cross plane (x) are slightly higher than in-plane (y-, y+). Out of field dose gradually decreases with off axis distance in all three planes. The out of field doses for various distance 5 cm up to 50 cm, and for different applicator from 6 x 6 cm2, 10 x 10 cm2, 14 x 14 cm2 at various depth 1, 3, 5, 10 cm. Outfield dose is higher in X-Plane as compared to Y plane due to X jaw replacement MLC.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-9: FEASIBILITY OF DYNAMIC FLAT BEAM TECHNIQUE IN FLATTENING FILTER FREE BEAM ONLY LINAC – COMPARISON OF DYNAMIC FLAT BEAM CHARACTERISTICS WITH STANDARD 6MV FLAT BEAM

S Thirumalai Swamy 1, G Arun 1, S Kala 1, A Anatharaman 1, Sai Subramanian 1

Introduction: In recent years application of flattening filter free beam (FFF) has increased tremendously due to its advantages like increased beam intensity, reduced head scatter & out-of-field dose and shorter beam-on time. Due to these advantages, linac vendors are now coming up with FFF beam only (Halycon). The one disadvantage of having only FFF beams is that for conventional 3D radiotherapy of large targets, total body irradiation and other extended SSD treatments such as mantle fields and spinal fields for crano-spinal irradiation, using FFF beams may not be possible. This issue can be addressed by including an internal optimization option to generate a flat beam profile but currently such an option is not available. Therefore in this study from a FFF beam, we try to create a dynamic flattened beam (DFB) using MLC's and IMRT optimization algorithm. Further, we analyze the beam parameters like flatness, symmetry, penumbra, full width at half maximum (FWHM) and cross compare it with standard 6MV flattened beam (FB).

Materials and Methods: In Varian Clinac-2100CD, beam parameters (gun current, voltage, etc.,) remain same for flattened and un-flattened beam, by removing the flattening filter from beam path increases the dose rate from 600 to 1400 MU/min and produces a forward peaked dose profile. Absence of flattening filter also increases the low energy X-ray component which leads to decrease in PDD from 66.6% (FB) to 63.8% (FFF). To create DFB, a homogeneous phantom image set (50 cm x 50 cm x 50 cm) was created in Eclipse TPS (V11.3). Dose calculations were performed in the above phantom for standard 6MV FB for different field sizes (FS) 6 cm x 6 cm, 10 cm x 10 cm, 16 cm x 16 cm and 20 cm x 20 cm and the dose was normalized at different depths of 1.5, 5, 10, 15 & 20 cm. Similar dose profiles were recreated for a 6MV FFF beam using DVO optimization algorithm and millennium 120 MLC. Further these dose profiles were exported to OmniPro software for beam profile analysis. Comparisons of beam parameters like flatness, symmetry, penumbra & FWHM were performed between the FB and DFB plans. Also the global hotspot (GHS), monitor units (MU) and gamma agreement index (GAI) were compared between both FB and DFB plans. Statistical analyses were performed using the Student's t-test (un-paired). In linac, beam profile measurement was performed using EPID for both FB & DFB.

Results and Discussion: The mean GHS was 165.1% (±58.3) and 172.6% (±67.4) for FB and DFB (p-0.7129). The increase in GHS in DFB was primarily due to difference in PDD. The mean MU was 160.4 (±57.4) and 385.7 (±206.4) for FB and DFB respectively (P < 0.0001). Due to the presence of non-uniform beam profile of the FFF beam, compared to the FB, the DFB beam tends to use more MUs/Modulation to deliver the uniform dose profile. The average GAI between FB and DFB was 98.5 (±0.32) with criteria 3 mm DTA and 3% dose difference. Flatness calculated for different FS and depth, the mean values of cross-line and inline values were 2.05 (±1.07), 2.29 (±1.16), 1.97 (±1.03) and 2.23 (±1.13) for FB and DFB respectively. Flatness for cross-line and inline of FB & DFB matches closely with p value of 0.5042 & 0.2507 respectively. This clearly indicates that dynamically modulated MLC can produce flat beam as good as flattening filter created flat beam. Similarly for symmetry, the mean value of cross-line and inline values were 0.22 (±0.09), 0.51 (±0.06), 0.19 (±0.12) and 0.51 (±0.30) for FB and DFB respectively (P - <0.0001 & <0.0001). Although this is statistically significant, the symmetry values were well within tolerance limit of 2%. For penumbra, mean values of cross-line and inline were 0.53 (±0.21), 0.55 (±0.19), 0.53 (±0.22) and 0.56 (±0.19) for FB and DFB respectively (P - <0.7461 & <0.561). For FWHM, the mean values of cross-line and inline values were 14.36 (±6.28), 14.31 (±6.19), 14.38 (±6.17) and 14.29 (±6.17) for FB and DFB respectively (P - 0.9828 & 0.9649). The results clearly show there is no significant difference in penumbra and FWHM between FB & DFB. Measured flatness, symmetry, penumbra & FWHM using EPID for 20 cm x 20 cm FS in FB were 1.09, 0.86, 0.34 & 20.2 (cross-line) and 1.52, 1.19, 0.3 & 20.2 (inline) respectively whereas for DFB the values were 1.22, 1.15, 0.36 & 20.1 and 1.45, 1.84, 0.3 & 20.2.

Conclusion: The result from this study shows that new type of linear accelerators with FFF beam only can still produce flattened beam with its beam characteristics similar to standard 6MV flat beam.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-10: CLINICAL EVALUATION FOR SETUP ACCURACY OF OSMS DATA OVER CBCT MATCHED VALUES FOR BRAIN METASTASIS WITH STEREOTACTIC RADIOSURGERY

Amol Bapu Pawar 1, Anand Jadhav 1, Ankita Nachankar 1, Prasad Dandekar 1

Introduction and Purpose: To evaluate the accuracy of a video-based optical surface imaging system for motion monitoring during stereotactic treatment of brain metastasis. Precise patient positioning and thus of the PTV is a prerequisite for effective treatment with SRS for brain metastasis. The intra fraction motion should at least be within the CTV-PTV margin used. Conventional imaging modalities used to ensure exact positioning for treatment typically involve additional radiation exposure of the patient. Patient alignment and monitoring during treatment, without additional exposure, is provided by optical 3D surface scanning and registration systems. Typical SRS brain treatments with multiple couch angels limits the ability of CBCT verification during treatment. This paper aims to correlate the data obtained by the OSMS system with the internal shifts observed by the offline CBCT matches pre and post treatment.

Materials and Methods: Patients treated with stereotactic radiosurgery (SRS) and fractionated stereotactic radiotherapy (FSRT) were immobilized with a thermoplastic double shell open face mask and a SRS immobilisation system (Encompass™ Immobilisation System and Encompass™ SRS Fiberplast Qfix). During treatment a video-based three-dimensional optical surface monitoring system was used to monitor the motion of a region of interest. This motion monitoring was done in 6 dimensions. A tolerance of 0.2 cm for linear directions and 0.5 degrees for rotational directions was set. If the optical surface monitoring system detected an exceeding of the set tolerance, treatment stopped automatically. If necessary, the physician decided to take a new CBCT. A total of 13 patients were followed for SRS or FSRT treatment between January 2016 and May 2018 with a total of 19 fractions evaluated for intra fractional uncertainties. Both CBCT and snapshots obtained with the OSMS were acquired at the start and stop of every treatment to compare both methods. In addition the average motion and SD during treatment was monitored to investigate the validity of pre- and post-measurements for assessing intra fractional motion.

Results: A pre-treatment mean intra fractional shift for of 0.12 cm in vertical (STDEV of 0.08), 0.11 cm in longitudinal (STDEV of 0.11), 0.17 in lateral direction (STDEV of 0.14). Only in one patient, set tolerance was exceeded and a new CBCT was taken which showed a lateral shift of 0.47 cm. A post-treatment mean intra fractional shift for of 0.04 cm in vertical (STDEV of 0.05), 0.05 cm in longitudinal (STDEV of 0.05), 0.04 in lateral direction (STDEV of 0.05). None of the patient were exceeded set tolerance during treatment.

Conclusion: The CBCT data clearly shows that the intra fractional offset of all brain metastasis patients treated with SRS or FSRT was below the institution's predefined threshold. The OSMS data obtained during treatment still needs a more detailed evaluation. For further analysis the approach was changed and real time data during treatment is now continuously triggered, obtained, mathematically analysed and compared with the CBCT offline calculated offset.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-12: INTENSITY MODULATED TREATMENT PLANNING FOR TOTAL BODY IRRADIATION

P Sabari Kumar 1, S Sathiyan 1

Introduction: TBI is traditionally treated using a conventional Linac using static AP/PA or parallel-opposed lateral beam arrangements at extended source-to-surface distance (SSD). Photon beam energies between 4 and 24MV can be used with tissue compensators to boost regions of varying patient thickness or shielding blocks to limits dose to organs at risk (lungs, kidney). Recently, there has been a shift toward more advanced TBI treatment technique with utilization of intensity modulation which achieves homogeneous dose distribution. The TBI treatment with IMRT doesn’t require the bigger bunker as well special equipments.

Objective: The aim of this study was to investigate the feasibility of achieving acceptable TBI plan using Intensity modulated technique with the Eclipse treatment system.

Methodology: A retrospective analysis was performed on four patient's CT data. CT images were obtained in the Head- First-Supine (HFS) orientation from the top of the skull to feet with a 5 mm slice thickness. In the case of greater length (more than 140 cm) patient treatment, another CT in Feet- First-Supine orientation is required. The FFS is intended for the legs to be treated with conventional AP/PA fields. A full body vacuum bag was used for immobilization. The planning target volume (PTV) was defined as the entire body, contracted to 2 mm below the skin. The planning aims were to deliver uniform 12Gy to the PTV while limiting the mean lung dose to less than 8Gy, and mean dose to kidney below 9Gy. A jagged-junction IMRT technique using multiple isocenter was used to overcome problems associated with field junctions and beam edge matching as well improves planning and treatment setup efficiencies with homogenous target dose distribution.

Results: The mean conformity index achieved was 0.93 ± 0.02 and the mean homogeneity index was 1.10 ± 0.02. The mean dose to PTV was 11.82Gy ± 0.2 and max dose (D2) of PTV was 12.62Gy ± 0.25. The dose for the OAR's were found to be within the acceptable limit, the lung mean dose was 7.89Gy ± 0.35 and the Kidney mean dose was 8.78Gy ± 0.48.

Discussion and Conclusion: This study has demonstrated the feasibility of achieving clinically acceptable IMRT TBI plans with Eclipse treatment planning system. The advantages of this technique include improved patient comfort and positioning reproducibility during treatment, accurate 3D dose information, and ability to selectively spare organs at risk. It is important to note that the irradiation of lower limbs was not included in this study. It is intended the legs be treated in the feet- first direction using AP/PA beams with conventional static fields. The Junction AP/PA beams for the legs are deemed acceptable owing to the absence of any organs at risk, and are the simplest approach to achieving the prescribed dose to the target.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-13: AN INDEPENDENT METHOD FOR MANUAL VERIFICATION OF MONITOR UNITS FOR MONOISOCENTRIC TREATMENT PLANS OF PATIENTS WITH CARCINOMA OF BREAST

Shrikant N Kale 1, S Hazarika 1, R Barman 1, R Kinhikar 1

Introduction: Radiation therapy (RT) in breast cancer has evolved dramatically over the past century. The CT-based 3-dimensional conformal radiotherapy (3DCRT) planning of Breast and Supraclavicular Fossa (SCF) can be achieved by either Dual Isocentric technique (DIT), one each for Breast and SCF or using single Isocentre for both Breast & SCF i.e. Mono-Isocentric technique (MIT). The AAPM Task Group report 71, recommends that it is an essential part of quality QA to verify Monitor units (MU's) to be delivered using treatment plan calculated by treatment planning system (TPS) with independent MU verification to correct any potential errors prior to the start of the treatment. In case of DIT, MU verification is established. But in case of MIT, the MU verification for breast plans becomes challenging due to involvement of quarter fields, blocked Isocentre, contour irregularity and complex treatment geometry.

Purpose: To establish a method for manual verification of the TPS calculated MU for MIT plans.

Materials and Methods: Twenty patients with MIT treatment plans were retrospectively selected for this study. All patients undergone CT based Simulation in supine position with appropriate immobilization devices and planned for 3DCRT MIT plans with optimizing photon energy, beam angles, beam weights, field in field technique with fixed collimator jaws using multileaf collimators(MLC) to achieve clinically acceptable plans. The dose calculation was done with calculation grid size of 2.5 mm using AAA algorithm in Eclipse TPS (v 13.5) with heterogeneity correction. As isocentre was blocked, the plans were normalized on reference points with prescription dose of 40 Gy in 15 fractions. The various parameters required for MU calculation were derived from TPS. The equivalent Square field sizes & water equivalent depths (WED) were determined from field Beams Eye View (BEV) and WED tool respectively. The corresponding output and TMR values were taken from lookup tables. The Effective target to Surface Distance (SSD) in reference point plane was determined to calculate inverse Square factor (ISF). The normalization factor for each individual field was determined at reference point by dividing percentage normalization value of the field by percentage normalization value of the plan. The doses for individual treatment fields were determined by multiplying normalization factor with prescription dose. The Enhanced Dynamic Wedge (EDW) factors for quarter field with both positive and negative off-axis distance (OAD) convention were calculated using EDW formula and verified dosimetrically as well as using Segmented Treatment Tables (STT) generated by treatment unit. The output factors were measured at centre points of quarter fields for all four quadrants. The output factors were also verified by TPS as well as from diagonal profiles used for beam configuration. This output factor was incorporated in the MU calculation to account for output of quarter field. All Dosimetric measurements were performed using 0.65 cc cylindrical ionization chamber using slab phantom. Finally the MU's were calculated using formula: MU = Dose (cGy)/Output (cGy/MU) * TMR * ISF * EDW (OAD) * Output factor.

Results: There is good agreement between TPS and manually calculated MU's, The Mean variation is found be 2 ± 1.9 MU.

Conclusion: A method for manual verification of the TPS calculated MU's for MIT plans has been devised and validated successfully. It is being routinely used for cross-checking the TPS MU's as an audit protocol in the department. As manual calculation do not consider all scatter factors & kernels like TPS, variation of ±5 MU is acceptable as a quality audit.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-14: COMPARISON OF ANISOTROPIC ANALYTICAL ALGORITHM AND ACUROS XB IN FLATTENING FILTER FREE BEAMS USING AN INHOMOGENOUS PHANTOM

S Vendhan 1, R Murali 1, N Arunai Nambiraj 1, M R Mariyappan 1, S Saraswathi Chitra 1, V Murali 1

Introduction: Flattening Filter Free (FFF) beams are widely used in volumetric modulated arc (VMAT) based Stereotactic Body Radiation Therapy (SBRT) of lung tumors. Different types of dose calculation algorithms are clinically available and employed to compute the dose distribution for SBRT plans. Presence of tissue in-homogeneity poses challenges in dose calculation and the accuracy of such calculated dose with FFF beams needs evaluation.

Objective: To validate and compare the accuracy of Anisotropic Analytical Algorithm (AAA) and Acuros XB dose calculation algorithms in flattening filter free (FFF) beams using an inhomogeneous phantom.

Materials and Methods: An inhomogeneous thorax phantom (Computerized Imaging Reference Systems Inc, USA) having dimension of 17.5 cm long x 30 cm wide x 20 cm thick were used for the study. The phantom has two distinct spaces having different electron densities characterizing human lungs surrounded by soft tissues. Both the regions have provisions for measurements using film and ion chamber along with suitable inserts. The thorax phantom was CT scanned with a slice thickness of 1mm and imported in Eclipse TPS (Varian Medical Systems, USA). Two different types of dose calculation algorithms, AAA and Acuros XB are available in Eclipse TPS. AAA is a model based algorithm which uses a 3D pencil beam convolution-superposition model to calculate the dose whereas Acuros XB utilizes a linear boltzmann transport equation for dose computation. Both the algorithms can be used for dose calculations with flattened and unflattened beams as well. For accuracy evaluation of algorithms at FFF beams in inhomogeneous media, treatment plans utilizing 6MV-FFF beams with varying degree of complexity which includes from simple open fields to complex modulated arc fields were planned in the thorax phantom. The doses were calculated in the phantom individually by both the algorithms using a calculation grid size of 2.5 mm for all the plans. Open field plans were created using jaws and MLCs separately for different field sizes (5, 10 and 15 cm2) for both static (AP-PA, bilateral and obliques) and arc (full and partial) delivery. 5 VMAT based SBRT plans were also planned for delivery and dose comparison between algorithms. The plans were then delivered in Truebeam SVC linear accelerator (Varian Medical Systems, USA) equipped with 120 Millennium Multileaf Collimator (MLC) with the thorax phantom and an ion chamber (0.125cc) at suitable place for measurements. The measured point doses were compared against the algorithm computed dose in TPS for all the plans. Before comparison, measured doses were corrected for daily output variation in linac.

Results and Discussion: For open static fields, comparison made between the measured and TPS computed dose shows a maximum dose difference of +0.85% and -0.4% for AAA and Acuros XB respectively. An average point dose difference of 0.55 ± 0.4% in AAA and 0.3 ± 0.1% in Acuros XB is observed for open arc fields across the field size studied. In VMAT plans, for AAA and Acuros XB, the maximum point dose deviation between measured and calculated is +1.7% and +1.1% respectively. Plans calculated with Acuros XB algorithm has shown better agreement with measured doses than AAA.

Conclusion: The result of our study shows that both AAA and Acuros XB dose calculation algorithms are comparable and accurate enough in dose computation at inhomogeneous media with FFF beams.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-15: A STUDY TO DETERMINE THE LEAST SUSCEPTIBLE PLANNING TECHNIQUE TO PROTON DOSE COMPUTATION ERROR IN PSEUDO CT

Gipson Joe Anto 1, S Harikrishna Etti 1, Bojarajan Perumal 1, Natarajan Ramar 1

Purpose: This is to study and determine IMPT planning technique, which has least susceptibility to CT number approximation of Pseudo CT images based on dose computation accuracy.

Materials and Methods: Five male pelvis Pseudo CTs were studied which were generated by Philips Ingenia MRI system known as MRCAT (Magnetic Resonance for Calculating ATtenuation) images. Pinnacle3 Proton treatment planning tool was used to generate treatment plans with Proteus Spot scanning proton machine. Regular CT image plans were used as reference plans. Steps of Study: (1) Multi field Optimization (MFO), Single Field Uniform Dose Optimization (SFUD), Robust Optimization (RO) plans were created in CT images of 5 patients for ca. prostate cases. (2) Created plans were copied to MRCAT images using Dynamic planning tool of Pinnacle3 along with ROIs and POIs. (3) Dose is computed in the MRCAT pseudo CT images with the same dose grid resolution of reference CT plans. (4) Published MRCAT CT to Density table and correspondingly scaled CT to stopping power were used for dose computation. (5) MRCAT plans are compared against corresponding reference CT plans. (6) MRCAT and CT plans were analyzed with DVHs, iso-dose lines and 3D gamma analysis.

Results and Discussion: Comparison of MFO plans shows significant difference in dose suggesting this planning technique is susceptible dose computation error in Pseudo CT images. In MFO plans dose modulation is high such that even a slight change in the geometric CT error would result in high dose error. SFUD plan evaluations suggests a reduction in dose difference (around 4% more points pass the 3D gamma) and seems relatively better planning technique for MRCAT plans. Robust optimization, which is supposed to make the plans less vulnerable to patient set up error and range error, seems a good choice as it smudges rapid dose fall and compensate for range error of proton beams. Around 8% more points pass gamma criteria for robustly optimized plans. Plans created by combination of robust optimization and SFUD also gives a robust plan with 8% hike in gamma passing points, stating it is less susceptible to dose error caused by stratified CT number approximation.

Conclusion: The dose error is mainly due to difference in the radiological distance (WET-Water Equivalent Thickness) of the data sets and error in spatial distribution of CT numbers. With the right CT to density table and CT to Stopping power table, error in radiological depth of the pseudo CT can be minimized. But the CT number geometric/spatial error is difficult to tame due to the high heterogeneity resolution of patient image data. This study shows Robust optimization with 5% range error input helps to reduce the sensitivity of dose to CT density approximation. Due to the difference in CT density distribution between CT and Pseudo CT images results in range error of spot scanning proton beams resulting in dose errors. These range errors can be minimized using robust optimization with range error margin. A combination of RO and SFUD also results in similar results but there is not enough evidence to state SFUD adds to the advantage of robust optimization.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-16: PERSONALIZED PLANNING WITH PINNACLE3 AUTO PLAN

Bojarajan Perumal 1, S Harikrishna Etti 1, R Vaitheeswaran 1, Natarajan Ramar 1, S R Meher 1, Gipson Joe Anto 1

Purpose: The purpose of this study is to investigate how the plan efficiency (quality) can be improved when Pinnacle Auto Plan is integrated with Plan IQ software.

Materials and Methods: Pinnacle3 Auto-Plan module is one of the established tools that can create clinically acceptable treatment plans without much intervention by the planner. On the other hand, Plan IQ is an effective tool designed to predict the achievability of clinical goals before even performing the optimization. In general, Plan IQ makes case-specific predictions whether a given clinical objective can be achieved or not. Because of this possibility given by Plan IQ, the user can define stringent goals for OARs and check their feasibility very quickly before even performing optimization. Hence, by using the Plan IQ suggested goals instead of using standard goals, it is possible to produce plans that are superior in terms of OAR sparing. These two tools (Auto Plan and Plan IQ) have already been integrated in Pinnacle in order to enable a seamless workflow experience to the users. In this study, our aim is to investigate how this integration helps improve the plan quality. We used Pinnacle3 version 16.2 to create clinical plans using Auto Plan tool coupled with Plan IQ for Prostate, Lung, and H&N cases. Basically we created two set of plans for each anatomy wherein one plan is created using standard clinical protocols such as RTOG and the other plan is created using Plan IQ suggested clinical goals.

Results and Discussion: We compared the dosimetric results for these two set of plans (RTOG driven Auto Plan à AP_RTOG and Plan IQ driven Auto Plan à AP_PlanIQ) for all three anatomies. We observed a significant reduction of OAR mean and max dose in AP_PlanIQ as compared to AP_RTOG without any compromise in target coverage. On the average, there is a reduction of OAR dose (Mean and Max) of about 39%, 6% and 21% for prostate, H&N and lung respectively in AP_PlanIQ plans as compared to AP_RTOG plans.

Conclusion: This study demonstrates that the Auto Plan tool can be efficiently used when integrated with Plan IQ. Also in many situations this integration helps reduce manual back tracking steps performed by the user when the user-fed goals are too stringent and potentially unachievable.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-17: A STUDY ON SELECTING OPTIMAL FLATTENING FILTER FREE BEAM QUALITY FOR INTRACRANIAL STEREOTACTIC RADIOTHERAPY

V Padmanaban 1, K Sollin Selvan 1, G Amirtharaj 1, V Srinivasan 1, Sunitha Prabhakar 1

Introduction: Use of FFF based Volumetric Modulated Arc Therapy (VMAT) for Hypo fractionated SRT (HSRT) is a well-established therapeutic modality for intracranial metastasis and benign lesions in the modern era. 6 MV-FFF and 10 MV-FFF beams available in Varian True beam STx linac offers high dose rate and increased dose per pulse which makes them an excellent choice for stereo tactic treatments. As HSRT employs large dose per fraction, energy selection criteria should incorporate the impact of high and low dose gradients, integral dose and the effect of the increased Monitor Units (MU).

Purpose: This study aims to critically analyze the selection of optimal beam quality for FFF-VMAT based intracranial SRT plans.

Materials and Methods: Fifteen intracranial SRT patients of different diagnose were studied retrospectively. The mean target volume was 10.46 ± 6.73 cm3 (Range: 1.37–23.96 cm3) with dose prescription of 18 Gy to 30 Gy in 3 to 5 fractions (mean ± SD: 21.7 ± 4.01). FFF based VMAT plans were created for both 6 MV-FFF (1400MU/min) and 10 MV-FFF (2400MU/min). An attempt was made to minimize the influence of other variables by forcing both the plans with identical target coverage. All the plans were devised with 3 arcs (2 non coplanar partial arcs and one coplanar full arc with same collimator and couch angles). For both the beam quality, optimizer was driven with the clinically accepted plan constraints and dose objectives with the single hold and synchronized time in each multi resolution (MR) levels with a single intermediate calculation using Eclipse TPS (v13.6) and dose calculations were performed using Acuros XB algorithm with calculation grid size of 1.25 mm. Statistical significance was assessed for all dosimetric parameters {GIHigh, GILow, CI, MU, HI, and Beam On Time (BOT)} and OAR doses using Wilcoxon signed rank test.

Results: 10 MV-FFF plans were found to have lesser MU compared to 6 MV-FFF (P < 0.002) plans. This effect is more pronounced in deep seated tumours. 10 MV-FFF resulted in very short BOT (P < 0.0003). No statistical significance were found in OAR doses (Brain stem (P < 0.054), optic nerve (P < 0.779), optic chiasm (P < 0.156), lens (P < 0.234) and cochlea (P < 0.28)) between two energies. Mean Paddick CI was 1.262 ± 0.13 and 1.257 ± 0.12 for 6 MV-FFF and 10 MV-FFF respectively. HI was found to be 1.381 ± 0.08 and 1.392 ± 0.09 for 6 MV-FFF and 10 MV-FFF. Both the CI and HI were not statistically significant. We found GIHigh (V90%PI/V50%PI) was 2.9% higher (P < 0.0006) for 10 MV-FFF and GILow (V25%PI/V50% PI) was -5.69% lower (P < 0.0006) in 10 MV-FFF compared to 6 MV-FFF plans. Results showed that low dose volumes V2Gy and V5Gy were statistically not significant with P values <0.0648 and <0.1902 respectively. High dose volumes (V10Gy and V12Gy) showed a significant increase in 10 MV-FFF with P values <0.00064 and <0.0008.

Discussion: The high dose rate (2400MU/min) available with 10 MV-FFF allows the gantry to maintain maximum speed even for a large dose per fraction which in turn reduces the treatment time. Increased PDD attributed to the sparse gradient in high dose region. Reduction in number of MU's reduces the integral dose and the risk of secondary malignancy. Normal brain tissue doses were comparable between these two plans which make the 10 MV-FFF as preferable energy.

Conclusion: Adequate importance should be given for the high dose gradient region (GIHigh) when 10 MV-FFF energy is selected for intracranial SRT especially when critical organs are nearby. Results showed that increased mean energy and high dose rate of 10 MV-FFF makes it an ideal choice for VMAT based intracranial SRT.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-18: EVALUATION OF OPTIMAL COMBINATION OF PLANNING PARAMETERS (FIELD WIDTH, PITCH, MODULATION FACTOR) IN HELICAL TOMOTHERAPY FOR BILATERAL BREAST CANCER

C A Muthuselvi 1, T K Bijina 1, B Subbulakshmi 1, A Pichandi 1

Introduction: Breast cancer is the most common malignancy among the women in the world, synchronous bilateral breast cancer is uncommon with the incidence of 2.1%. Bilateral Breast planning is time consuming and challenging because of the huge volume and nearby critical structures. Helical Tomotherapy (HT) is capable to deliver well tolerated homogeneous dose to bilateral breast without field overlapping.

Purpose: Aim of this study is to evaluate the influence of HT treatment planning parameters on plan quality and treatment time for bilateral breast and to find the optimal combination of planning parameters.

Materials and Methods: We have evaluated 5 patients CT data sets with 90 plans. For each patient, 18 plans were created using the combination of planning parameters (Field width (FW) of 2.5 cm, 5 cm; Pitch of 0.215, 0.287, 0.43; Modulation Factor (MF) of 2, 2.5, 3). For every patient initial plan was created with FW 5cm, Pitch 0.287, MF 2.5 and the PTV was prescribed to 50Gy in 25 fractions. Using helping structure we have blocked the beam from posterior direction to reduce the low dose spillage. We have optimized the plan to achieve 50Gy to 95% of the PTV, without increasing 107% dose more than 2% volume. After achieving acceptable OAR results, this plan was copied with its optimization constrain and 17 more plans were created by changing only its plan parameters. Plans were evaluated by dose volume histogram (DVH) analysis. Plan quality of target was quantified using Homogeneity Index (HI), Conformity Index (CI), Dmin by D98%, Dmax by D2%, and coverage by D95%. Organ at risk (OAR) doses were evaluated by mean dose, V5Gy, V25Gy for heart and mean dose, V5Gy, V20Gy for both the lungs. Treatment time were also evaluated for all the 90 plans.

Results: PTV: When FW lowered from 5 cm to 2.5 cm the CI and HI of PTV improved from 0.997 to 0.999 and 0.07 to 0.04. HI decrease with (more homogeneous) decreasing the pitch or increasing modulation factor. The average D98% and D2% were 49.2Gy, 49.4Gy, 49.5Gy and 51.8Gy, 52.1Gy, 53.2Gy for pitch value 0.215, 0.287, 0.43 respectively. Increasing MF slightly improved all the PTV indices. OAR: Average value of heart and lung mean doses were 4.89Gy, 5.17Gy and 10.5Gy, 10.7Gy for field width 2.5 cm and 5 cm respectively. V5Gy of, heart was 21.4%, 23.1%, 24.8% and lung was 45.6Gy, 46.7Gy, 47.8Gy for pitch value 0.215, 0.287 and 0.43. Increasing MF improved all the OAR indices. Treatment Time: As expected FW of 2.5 cm (~10 min) had a higher treatment time than 5 cm (~6 min). Pitch value didn’t affect the treatment time. Increasing MF increased treatment time by 2-3 mins.

Discussion and Conclusion: Comparison of dosimetric indices showed that FW 2.5 cm improved all the indices but increased treatment time 40–50% than 5 cm FW. Pitch value 0.43 didn’t offer any dosimetric advantage. Among the compared plans pitch value 0.215 showed better results compared to 0.287 and 0.43 for OAR dose (mean, V5Gy, V25Gy for heart and mean, V5Gy, V20Gy for both the lungs). Increasing MF increased plan quality as well as treatment time. By applying small FW, tighter pitch and large MF values, it is possible to get a sophisticated treatment plan with a very long treatment time. However, this results in two adverse outcomes: patient discomfort (to lie down static during irradiation) and inherent organ movement due to breathing. Considering all these and on the basis of our analysis, plan with FW 5 cm, pitch 0.215 and MF 2.5 can be considered as a optimal combination of planning parameters for bilateral breast irradiation in HT technique.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-19: EVALUATION OF DEFORMABLE IMAGE REGISTRATION AND DOSE ACCUMULATION FOR PROSTATE STEREOTACTIC BODY RADIOTHERAPY PATIENTS

Swamidas V Jamema 1, Reena Phurailatpam 1, Subhajit Panda 1, Vedang Murthy 1, Kishore Joshi 1, D D Deshpande 1

Aim: To quantify the difference between planned and the delivered dose using deformable image registration (DIR) and deformable dose accumulation (DDA) for patients treated with Stereotactic Body Radiotherapy (SBRT) for prostate cancer.

Materials and Methods: Five prostate cancer patients previously treated with SBRT (35Gy in 5 fractions) were retrospectively analysed for this study. All patients were treated with Rapid Arc (RA) using 10MV FFF beams (Varian Eclipse TPS v13.5.37, True Beam v 2.1, Varian Medical Systems, Palo Alto). Daily cone beam CT (CBCT) was carried out as per the institutional protocol, which includes a bladder filling and a bowel protocol which reproduces the anatomy on the treatment day as compared to the planning day. CBCT was acquired for a full FOV that includes PTV, CTV, bladder rectum, and part of the bowel. Delivered dose was calculated by accumulating the doses using DIR and DDA in Velocity DIR software (v 3.2.1). Planning CT images (pCT), RT structures, RT plan and RT dose were imported from the planning system to the DIR system. Daily CBCTs were also imported and deformably registered to the pCT in DIR software using CBCT corrected DIR algorithm and generated a Synthetic CT (sCT). RT plan was recalculated on sCTs. The dose calculated on sCTs were deformed back and accumulated onto pCT to estimate the total delivered dose. The planned and delivered doses were compared using various dose volume parameters for PTV, CTV and OARs (bladder and rectum). Target coverage was estimated by comparing the differences between the planned and accumulated doses of PTV and CTV in terms of V95% (%), Dmean (Gy) and Dmax (Gy). For OARs, Dmax (Gy), Dmean (Gy), volume receiving various doses such as 30Gy, 20Gy, 10Gy and 5Gy were evaluated. Registration accuracy were also evaluated using Dice Similarity Co-efficient (DSC) and Hausdorff distances (Hf avg).

Results: Mean (±sd) volume of PTV, CTV of pCT was 54 (±9)cc and 26 (±7)cc respectively. Mean (±sd) volume of bladder and rectum of pCT vs CBCT averaged over 5 fractions were 365 (±93) vs 344 (±79)cc and 44 (±6) vs 45 (±6)cc respectively. Mean (±sd) of DSC of bladder and rectum were 0.89 (±0.7) and 0.9 (±0.07), while Hf avg bladder and rectum were 2.8 (1.3) mm and 1.1 (0.9) mm respectively. Mean (±sd) difference in V95% between the planned and the delivered dose for PTV was 3.4 (±5)%. However, maximum variation in V95% between the delivered and the planned dose was found to be correlating with the variation in bladder volume on that particular treatment day. For example, the variation in patient 1, 2, 3, 4, 5 along with their bladder volume variation is as follows: 9.9% (138 cc), -8.5% (114 cc), -17% (109 cc), -3% (14 cc) and 7.9% (36cc). Three out of four patients who had dose variation of more than 5% had bladder volume variation of >100cc. However the variation in other dose volume parameters was found to be negligible as well as for CTV. Mean (sd) difference between the planned vs delivered dose for bladder was -0.07 (1.5)%. However, patient wise analysis showed a maximum variation of 19% on the day when the bladder volume variation was 138cc (patient-1). We had also observed that the difference between the planned vs delivered dose for other patients was also correlating with the bladder volume variation. Mean (sd) difference between the planned vs delivered dose for rectum was 0.8 (±2)%. However patient wise analysis showed a maximum variation of 3.7%. In rectum the maximum volume variation was 12cc, and no correlation of dose variation with rectal volume variation was observed.

Conclusions: Deformable dose accumulation of SBRT prostate patients was feasible. The difference between the planned and the delivered dose to target structures and OARs were quantified.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-20: EVALUATION OF KNOWLEDGE BASED RAPID PLAN MODEL FOR PATIENTS OF CARCINOMA CERVIX

H Mahendran 1, Deepa Thekedath 1

Introduction: Cancer is one of the most complex diseases and one of the most effective treatments, Radiation therapy, is also a complicated process. Informatics is becoming a critical tool for clinician, Medical Physicists, scientists for improvement to the treatment and better understanding of the diseases. Computation techniques such as Knowledge based planning/Artificial Intelligence have been increasingly used in radiation therapy. Artificially intelligent computer systems are hungry for data. As a result, they must be continuously fed with the right kind of data so they continuously learn and update their algorithms. As AI is incorporated into cancer treatment planning systems, the impact of automation on existing workforce skill sets must be evaluated.

Purpose: This study is to evaluate the feasibility of using a knowledge based Rapid plan model to generate new intensity modulated radiation therapy plan for carcinoma cervix patients.

Methods: The 102 consecutive patients treated in our centre for Carcinoma Cervix were taken for this study. The above said patients are treated using the Intensity modulated radiation therapy (step and shoot method) planned in Eclipse treatment planning system (version 13.7). Those patients DVH data were extracted to generate Knowledge based Rapid plan model. The retrospective study were performed in Rapid plan model for twenty patients. Original IMRT plan mean OAR (Organ at Risk) doses compared with mean doses achieved when Rapid plan is used to make the new plan. Difference between the achieved and predicted DVH curves was analyzed. Finally the rapid plan prediction are used to evaluate achieved OAR sparing of Automatic Interactively optimized plan (AIO) and Manually interactively Optimized Plans.

Results: The Plan analysis performed based on dose distribution and Dose Volume Histogram for Planning target volume (PTV), the organ at risk (OARs) for bladder, rectum, small bowel, right and left femoral heads as well as other physical indices like Mean dose (Dmean), Maximum dose (Dmax), 95% dose (D95), %5 dose (D5), total number of segments, monitor units, Dose homogeneity Index and Conformity Index for the PTV were also evaluated.

Discussion: For all twenty patients with help of Rapid plan model we achieved a clinically acceptable plans and the same was compared with original plans. Dose homogeneity index in both plan is 1.03 ± 0.01 and significance difference P = 0.02. Conformity index of PTV is 1.08 ± 0.01 for both plans and significance difference of CI P = 0.002. Similarly OAR doses for bladder, rectum, bowel, Right femoral head and Left femoral head were evaluated there is no much variation in both plans for all twenty patients the significance difference for all critical organs (P < 0.0001).

Conclusion: We conclude that knowledge based Rapid plan is truly automated planning system. It generate clinically acceptable treatment plans of high quality in less time duration. This automated approach can improve the efficiency of the treatment planning process by ensuring that quality plans are developed for treatment.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-21: EVALUATION OF THE INTENSITY MODULATED RADIOTHERAPY TREATMENT PLANS COMPUTED USING PHYSICAL AND BIOLOGICAL OPTIMIZATION ALGORITHM

Suvankar Das 1, I Rabi Raja Singh 1

Introduction: The main goal of radiation therapy is to deliver adequate high dose to the tumor so that all tumor cells are killed while avoiding the radiation induced damage to the surrounding normal tissue. Radiobiological models are used in the radiobiological treatment planning to estimate the tumor control probability (TCP) and the normal tissue complication probability (NTCP). TCP-PoissonLQ model is used to estimate the TCP and two models namely NTCP-PoissonLQ and NTCP-LKB model are used for the computation of NTCP. The TCP-PoissonLQ model and the NTCP-PoissonLQ model are based on the cell survival model and the Poisson statistics. The NTCP-LKB model is based on the probate function.

Purpose: (1) Evaluation of the Intensity Modulated Radiotherapy Plans using physical and biological evaluation tool. (2) Generation of IMRT plans using biological optimization algorithm in terms of probability for Complication Free Tumor Control, Tumor Control Probability and Normal Tissue Complication Probability. (3) Comparison of IMRT plans computed using physical optimization algorithm and the same using biological optimization algorithm.

Materials and Methods: In this study we have used the Eclipse Treatment Planning System Version 13.7 (Varian Medical Systems, Inc., USA). It uses Anisotropic Analytical calculation algorithm (AAA) with a calculation grid of 2.5 mm for dose calculation. It has both physical and biological optimization algorithms for the generation of treatment plans and uses Dose Volume (DV) as well as Biological evaluation tools to evaluate the plans. Biological based evaluation tools are developed by the Ray Search Laboratories (Suveavagen25 111 34 Stockholm, Sweden). 20 numbers of patients with Head and Neck cancers treated with Intensity Modulated Radiotherapy technique were selected. Treatment plans of these cases were already performed by the AAA algorithm using a calculation grid of 2.5 mm. In each plan, the physical parameters such as maximum, minimum and mean dose for both Planning Target Volume (PTV) and Organ At Risk (OAR) were obtained from the Dose –Volume Histogram (DVH) generated. At first, these plans were evaluated by using Biological Evaluation tool. Secondly these same patients were re-planned using Biological based Optimization algorithm which calculates the biological parameters such as TCP and NTCP. Results were obtained in terms of probability for complication free tumor control, Tumor Control Probability and Normal Tissue Complication Probability. The ‘t-test’ was carried out to find out whether the values of P+, TCP and NTCP have significantly changed for both physical optimization and the biological optimization.

Results: From the ‘t-test’ values it was found that the ‘P’ values obtained from the comparison for P+, TCP and NTCP were 0.001, 0.034, 0.04 respectively. Since these values are less than 0.05 we conclude that there is a significant change in the result of biological optimization when compared to that obtained from physical optimization. The average value of P+ for the Physical Optimization was 14.58 and for Biological Optimization was 32.34. Hence complication free tumor control has increased when Biological Optimization is used. The average value of TCP for the Physical Optimization was 38.18 and that for the Biological Optimization was 62.47. This shows the probability for tumor control has increased when Biological Optimization is used. The average value of NTCP for the Physical Optimization was 54.75 and the same for Biological Optimization was 46.10. Hence the probability of normal tissue complication is relatively less when Biological optimization is used.

Discussion and Conclusion: It is evident from this study that there is a significant increase in the probability of complication frees tumor control as well as Tumor Control Probability anticipated when we use the biological optimization. There is also significant decrease found in the probability of normal tissue complication while we used biological optimization. Hence it is concluded that the biological optimization is necessary to enhance the complication free tumor control while reducing the normal tissue complication.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-22: COMPREHENSIVE QUALITY ASSURANCE OF INTENSITY MODULATED RADIOTHERAPY AND VOLUMETRIC MODULATED ARC THERAPY TREATMENT USING ARCCHECK DIODE ARRAY, PORTAL DOSIMETRY AND LINAC TRAJECTORY LOG FILE ANALYSIS

Prashantkumar Shinde 1, Mukesh Meshram 1

Introduction: In modern Radiotherapy, IMRT (Intensity Modulated Radiotherapy) and VMAT (Volumetric Modulated Arc Therapy) treatment delivery techniques are routinely practised. These are most complex due to various intricate delivery mechanism, sequencing algorithms and calculation algorithms. Thus verification of IMRT and VMAT treatment delivery plays a paramount role in clinical practice to evaluate the quality of radiotherapy treatment plan. For this a comprehensive quality assurance programme for patient treatment verification is important to assess the accuracy in machine delivery as per the desired accuracy.

Purpose: To compare and evaluate three different dosimetry system; ArcCHECKTM Diode Array, aSi-1200 EPID Portal Dosimetry and Trajectory log file Analysis for comprehensive quality assurance of IMRT and VMAT treatment delivery.

Materials and Methods: This study consist of assessment of Patient specific treatment plan quality assurance using ArcCHECKTM Diode array, VarianTM (amorphous silicon based) aSi-1200 EPID based Portal Dosimetry and Treatment Trajectory log file analysis of thirty patient who underwent IMRT and VMAT treatment on Varian TrueBeamTM Linear Accelerator. All Treatment plan were calculated using AAA Algorithm in EclipseTM (v13.7) Treatment planning system. Trajectory log files are generated in TrueBeamTM delivery and record system after delivery of the treatment plan, which records normalized MU delivered, Angle of gantry and Actual and expected position of each leaf of MLC in either Bank of MLC at every 20 millisecond. These trajectory log files generated in TrueBeamTM after performing Quality assurance of Treatment plan using ArcCHECKTM and aSi-1200 EPID portal dosimetry. These Trajectory log files were analysed using Assurance QA software. Gamma Analysis metric of DTA (Distance to Agreement) 3 mm & 3% with threshold dose of 10% is used for Gamma Analysis of TPS and Measured fluence comparison. Comparative evaluation of ArcCHECKTM Dosimetry, Portal Dosimetry and Trajectory log file analysis for thirty QA plans were done.

Results: The average Gamma pass rate (%) (γ ≤ 1) of QA plans for ArcCHECKTM, a-Si 1200 EPID portal Dosimetry and Trajectory Log file analysis are 99.05 ± 1.04, 98.38 ± 1.63 and 99.93 ± 0.08 respectively. The average dose (%) difference (γ > 1) in TPS and delivered plan is 0.95 ± 1.04, 1.62 ± 1.63 and 0.07 ± 0.08 respectively in ArcCHECKTM, Portal Dosimetry and Trajectory log file analysis. Mean Average Gamma (γ) for ArcCHECKTM, Portal dosimetry and Trajectory log file analysis is 0.37 ± 0.09, 0.38 ± 0.05 and 0.08 ± 0.08 respectively. Mean Maximum dose variation (γmax) in calculated plan and delivered plan for ArcCHECKTM, portal dosimetry and Trajectory log file analysis is 1.37 ± 0.44, 1.81 ± 0.84 and 0.46 ± 0.43 respectively.

Discussion: The average gamma pass rate (γ ≤ 1) in ArcCHECKTM, portal dosimetry and Trajectory log file analysis is highly comparable to each other and constitute a comprehensive Quality assurance tool. Average dose difference (γ >1) is more in ArcCHECKTM and portal dosimetry compare to Trajectory log file Gamma analysis. This is due to more complex parameter that influence the accuracy and stability of ArcCHECKTM diode detector and a-Si detectors in EPID.

Conclusion: Comparable results of Gamma Analysis for ArcCHECKTM, Portal Dosimetry and Trajectory Log File shows comprehensive and robust Quality of TrueBeamTM Delivery System. Portal Dosimetry can be used as a quick and efficient quality check system. Trajectory log file Analysis saves time for QA setup and is very good tool for quick and reliable dosimetry verification for Patient specific Plan quality assurance.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-23: WIRELESS ROBOTIC PHANTOM FOR QUALITY ASSURANCE OF 4D GATED RADIATION THERAPY

J Sujith Christopher 1, Paul B Ravindran 1

Background: High precision Radiotherapy Technique such as Intensity Modulated Radiotherapy (IMRT), Volumetric Arc Therapy (VMAT) and Three-Dimensional Conformal Radiotherapy (3DCRT) requires accurate targeting of Tumour. However, due to cardiac and respiratory motions, achieving precise targeting is ambiguous in cases of intra-thoracic tumours, due to varying magnitude of displacement with respect to respiratory pattern and size of tumour of the patient. These displacement results in an increased treated volume and distribution of the dose delivered does not match with the intended dose distribution.

Purpose: To evaluate the accuracy of radiation dose delivery during gated radiation therapy with the Varian Realtime Position Management (RPM) System using an in house developed wireless respiratory phantom.

Materials and Methods: The Respiratory phantom was constructed and evaluated by Nihal Daniel and Paul Ravindran in 2009. However, in the present study we have improvised the control and simulation of respiratory phantom by introducing wireless communication through an Android mobile Application from the console room. The phantom is battery powered which eliminates the use of very long data transferring cables and power cables. The wireless communication is configured through two HC-12 Bluetooth modules connected as a serial transponder since the radiofrequency signals cannot pernitrate the wall thickness of Linac. Two Arduino Nano microcontrollers connected with the transponders which acts as the control unit for the respiratory phantom on the couch and for the Android application communication from the console. The unit in placed in the Console is configured with an additional HM-10 Bluetooth module for pairing up with the Android device. The movement of the Respiratory phantom is simulated through powerful stepper motor arranged in the setup. The control unit can move the phantom in x and y-axis and with varying frequency in accordance with the recorded RPM pattern. The respiratory pattern from RPM is taken and fed into the control unit of the phantom to simulate the recorded respiratory movement of the patient.

Results: The validation of simulation is done by making the phantom to mimic the recorded respiratory pattern of a patient in which the movement of the phantom is again monitored using RPM and verified against the actual plot. The DIBH was also simulated with the phantom for the automatic gating system to turn the treatment beam on and off.

Discussion: Thus, the wireless robotic respiratory phantom is very useful for doing a patient specific quality assurance in Gated treatment delivery. Since no cables are involved and the phantom is battery powered it's easy to setup. The use of Android application for controlling the phantom eliminates simulation errors caused due to voltage fluctuation in conventional electronics control unit.

Conclusion: The Respiratory phantom has been found to be useful in performing patient specific QA and accessing the effectiveness of the RPM gating system. The Varian RPM gating is found be effective in delivering gated Radiotherapy.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-24: NOVEL TOOL FOR MODELLING MECHANICAL OFFSETS WITH RADIATION ISOCENTRE AND EVALUATE THE EFFECT OF CLINICAL SRS/SRT TREATMENT

Nirmal Babu 1, M Chandrasekaran 1, D Johnson 1, S McInally 1, A Naga 1, N Khater 1, A Blackmore 1, C Naish 1, C Birch 1

Aim: To analyse the impact of mechanical offsets of gantry, couch and collimator with radiation isocentre using an in-house fabricated tool on SRS/SRT treatment plan quality parameters including target coverage, selectivity index (SI), conformity index (CI), gradient index (GI), dose to GTV, PTV and OARs.

Materials and Methods: Tool: An in house fabricated tool was used to locate the radiation isocentre with the precision of micro mm with 3 independent micro meter in X, Y, Z direction. The tool consists of a metal ball bearing of radius 5 mm placed at the geometric centre of a hollow fibre box (22 x 22x 22cm). The whole box ball bearing assembly is mounted on a fixation system along with the micrometres to enable translational movements and a rotation screw mechanism applied rotations to cancel out yaw, pitch and roll errors. The box tool was positioned on the couch followed by zeroing of any rotational errors using a spirit level.

Pre Verification: Four radio opaque markers per face of the box that are permanently fixed with a known geometry were portal imaged with gantry 0 and 90 for a field size of 10x10 cm2 to check the alignment of the ball bearing centre with box without any tilt.

Radiation Isocentre: The MV radiation isocentre of an unflattened 6MV beam of an Elekta Agility linac was identified by acquiring images at 4 cardinal and 4 oblique gantry angles for a field size of 3x3 cm2 using iviewGT portal imaging device. The images were analysed using Pipspro software. Any deviation in the position of the geometrical centre of the ball bearing to that of the radiation isocentre was corrected by applying shifts calculated by pipspro using the micrometers attached to the tool. The test was repeated until the geometrical centre of the ball bearing matched the mean of the radiation isocentre.

Laser Setup: The box had slits of 2 mm cut out in ‘plus’ shape at the centre of the left, right and anterior faces which enabled readjustment of the lateral, sagittal and overhead lasers to match the mean of the radiation isocentre of the linac.

Mechanical Isocentre: A calibrated mechanical front pointer was fixed on the gantry such that the pointer tip matches the laser centre. An extended leaning pointer assembly was attached in place of the ball bearing. The pointer can be also moved with the same XYZ micrometer assembly. Radiation isocentre coordinates from the XYZ scale were noted for every 45 deg gantry angle, by matching the learning pointer tip to the front pointer by moving the independent XYZ scale. The difference in the scale value with respect to radiation isocentre coordinate is the actual mechanical shift from the radiation isocentre. The same method was followed for various couch and collimator angles. From these offset values the mechanical offset of the machine was plotted with respect to gantry, couch and collimator. Two models were created, an actual error model which is pre adjusted the radiation isocentre and corrected model which is adjusted the radiation isocentre to fit into the mechanical offset.

Clinical Analysis: Plans were created on an anthropomorphic head phantom on Pinnacle TPS v.16.0 as per our local SRS clinical protocol based on 2 clinical plans. These original plans were recalculated with the same beam parameters (beam shape, MUs) but with a new offset isocentre which was derived from the 2 models. The corrected plans and original plans were compared and plan parameters were analysed for both the models.

Results and Discussion: A Novel tool was successfully used for:

  1. Effectively locating the radiation isocentre with micro meter accuracy.

  2. In-Room laser alignment to the radiation centre.

  3. Modelling mechanical Offset was modelled by plotting the offset of radiation centre with respect to gantry, couch and collimator.

Collimator and gantry-lateral offsets were found to be very small (max 0.2mm) and were not considered in clinical correction. On the clinical plan comparison, GTV dose has no variation on both the models and PTV coverage was noted with a maximum of 1 % volume deviation and better with corrected model where the radiation isocentre corrected to mechanical centre and OAR doses found a minimal variation but within our clinical tolerance.

This tool can be used for routine linac QA, acceptance and quicker laser setup.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-25: AUTOMATED TREATMENT PLAN QUALITY ASSURANCE USING ECLIPSE SCRIPTING APPLICATION PROGRAMMING INTERFACE

P Jenifer Sneha 1, B Paul Ravindran 1

Introduction: The main objective of this work is to estimate the effectiveness of Plan Quality Analyser (PQA) Script which was developed to estimate plan quality, reduce human error, increase the efficiency of electronic workflow, standardize and automate the plan review in the treatment planning system (TPS). PQA Script is used as an interface to compare the data from TPS and the ARIA Database. The fundamental aim behind PQA Script was to reduce the amount of time spent in reviewing the plan parameters for each plan and enhance the time for validating the dosimetric plan quality.

Purpose: In order to improve safety, radiation oncology departments have deployed checklists to standardize processes. It has been seen that the review check of the treatment plan has the greatest effectiveness for catching errors. Manuel checking of plan parameters can lead the physicist to focus on the mundane details instead of overall plan quality, and increase the chances of missing serious errors. The use of automation will bring more effect to safety and quality. Since the plan evaluation involves checking similar plan parameters for each patient, this process can be automated.

Materials and Methods: External beam treatment planning is performed with Eclipse (Version 13.7), the PQA Script was developed using C# in Visual Studio 2013. Eclipse Scripting Application Programming Interface (ESAPI) is the programming interface in Eclipse. PQA Script can perform Quality Assurance (QA) check for all types of treatment plans such as Volumetric modulated arc radiotherapy (VMAT),

Intensity modulated radiotherapy (IMRT), Three-dimensional conformal radiotherapy (3D CRT). The PQA Script verifies the plan for 1. User origin position, 2. Inclusion of correct couch structure, 3. Complimentary collimator angle for VMAT plan, 4. Jaw tracking inclusion, 5. Field size, 6. Isocentre position of beam and a few other parameters that are necessary to be checked. PQA Script is made machine specific for better performance. The PQA script includes a Dose Volume Histogram (DVH) Analyser. Physicists can choose the desired Protocol before performing DVH analysis. One vital reason for the development of the PQA Script is to eradicate human error. PQA Script checks for Patient name, appropriate Image, Prescribed Dose, Prescribed Fraction and makes an alert to the physicist in case of any error.

Results: In this work, the script is run from the Eclipse Treatment Planning System tool bar which automates the complete Quality Assurance check and generates a file that is to be documented. The file is successfully generated for the plan where the script has not encounter any error. The Script is used clinically as an electronic checklist before finalizing the plan. The PQA script along with DVH evaluation for a plan is tested and practised clinically.

Discussion and Conclusion: The PQA script eliminates human error in the treatment plan that could have significant effect on patient treatment and also helps to analyse the DVH quickly. PQA Script being, an electronic process of checking treatment plans has made the cumbersome efforts of physicist simple and less time consuming.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-26: DOSIMETRIC COMPARISON OF GRAPHICAL AND INVERSE PLANNING SIMULATED ANNEALING BASED DOSE OPTIMIZATIONS PLANS WITH VARYING VALUES OF DWELL TIME DEVIATION CONSTRAINTS FOR INTERSTITIAL BRACHYTHERAPY OF CANCER CERVIX

Saurabh Roy 1, V Subramani 1, Kishore Singh 1, A K Rathi 1, Savita Arora 1, Aditi Aggarwal 1

Introduction: Inverse planning in brachytherapy is a new concept in which very few institutes are venturing. Initial applications of inverse planning simulated annealing (IPSA) for treatment of prostate cases have yielded encouraging results. But further involvement of IPSA in gynaecological sites is not being done. Graphical optimization in interstitial application of cancer cervix is common practise for achieving therapeutic goals. The purpose of this retrospective study is to explore new possibilities of application of IPSA in interstitial brachytherapy of cancer cervix. Large deviation in dwell time while using IPSA optimization can be controlled by applying dwell time deviation constraints (DTDC). In some interstitial brachytherapy plans of cancer cervix based on the graphical optimization do not give sufficient target coverage while retaining the dose to organs at risk within recommended limits. In this study we examined whether such plans could be improved using IPSA with varying values of DTDC parameter. This is one of the very first studies in which DTDC applied IPSA plans are compared with graphically optimized interstitial brachytherapy of cancer cervix.

Purpose: The aim of this retrospective study is to implement IPSA optimized plans for interstitial brachytherapy of cancer cervix with varying values of DTDC parameter and compare the results with graphically optimized plans.

Materials and Methods: For this retrospective study we have generated IPSA optimized interstitial brachytherapy plans for 20 patients by using Oncentra Brachytherapy treatment planning system (version 4.3). The DTDC value of IPSA plan was increased from 0.0 to 1.0 in step of 0.2. The variation in dose volume histogram parameters D90, V100, V150 and V200, for CTV were studied and compared with graphically optimized plans. For dose analysis of bladder and rectum, their D2cc parameter was investigated. The conformity index (CI) and homogeneity index (HI), were also computed.

Results: Good target coverage for prescription dose was achieved with graphically optimized plans as compared with IPSA optimized plans. Homogeneity was good with the IPSA-based technique as compared with the graphically optimized plans. The homogeneity index was increasing with increasing value of DTDC. Graphically optimized plans resulted in a higher CI (with a mean value of 0.83) compared with the IPSA based optimization of DTDC value 0 (with a mean value of 0.71). But increasing DTDC value of IPSA plans resulted in higher CI compared with lower DTDC valued IPSA plans. V150 and V200 values were lower for DTDC based IPSA plans compared with graphically optimized plans. D2cc values for rectum and bladder were lower for graphically optimized plans compared with IPSA based plans.

Discussion: Adequate target coverage was achieved by graphically optimised plans compared with IPSA plans of varying values of DTDC. A higher value of DTDC in IPSA plan removes the larger of the dwell time in isolated dwell positions and also reduces overall treatment time, but this is at the cost of decreasing D90% and V100% value which indicate lower CTV coverage. Lower values of V150 and V200 for DTDC based IPSA plans indicate lower high dose volumes compared with graphically optimized plans. Greater sparing of bladder and rectum were observed in graphically optimised plans compared to IPSA plans with any value of DTDC.

Conclusion: This study shows that, graphical optimization may achieve good target coverage for interstitial brachytherapy of cancer cervix but DTDC based IPSA plans restrict the occurrences of high dose volumes.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-27: PERFORMANCE EVALUATION OF AXXENT ELECTRONIC BRACHYTHERAPY SYSTEM MODEL 110

Smriti Sharma 1, Namitha Krishnakumar 1, S Mahalakshmi 1, G Sahani 1, P K Dash Sharma 1

Introduction: Traditional brachytherapy refers to the placement of radioactive sources on or inside the tumor. Electronic brachytherapy (EB) is new form of radiotherapy which utilizes miniature X-ray source instead of radioactive source to deliver desired doses to tumor. The advantages of EB include low dose to surrounding organs, simple emergency handling procedures, lesser shielding requirements, no leakage radiation in off state and no radioactive waste. Purpose of this paper is to discuss performance evaluation of Axxent EB system model 110 with respect to conformance with manufacturer's specifications and existing QA protocol of brachytherapy, as applicable. This unit is intended to deliver intracavitary, surface radiation and Intra Operative Radiation Therapy(IORT) and not suitable for interstitial application due to larger source dimension.

Materials and Methods: The Axxent EB system model 110 is manufactured by M/s. XOFT (a Subsidiary of iCAD), USA. Device contains a miniature X-ray source, a controller, cooling tube set and applicators. The Controller supplies high voltage, filament current, and circulates cooling water to the X-ray Source. The dimension of X-ray tube is 1.5mm (dia) X 3.0 mm (length). The operating voltage and current are 50kV and 300μA respectively. X-ray source moves in stepped manner with minimum step size 1.0mm and maximum treatment length 7.5cm. The radiation output of the source is determined using the AAPM Task Group 43 formalism, using well chamber and electrometer attached to the unit. The equipment is designed to check several self-tests including constancy of air kerma rate prior to each treatment. As no national/international protocol is available, the existing QA protocol of brachytherapy, as applicable, was taken as base document along with manufacturer's specifications for testing purposes. Testing of the engineered safety features of the equipment and dosimetric tests were carried out. Some of tests includes interlocks such as emergency stop, pullback arm lock, check of cooling water level, coupling of the X-ray source with treatment unit, coupling of the X-ray source with pullback arm nest, coupling of the applicator with controller arm. Other tests are source position accuracy and reproducibility, timer linearity and error, and tests for treatment planning system. Tests results related to emergency and failure such as failure of controlling timer, positional accuracy and movement of sources were provided by manufacturer. In this model control console is located on the brachytherapy unit and operated from inside the treatment room. Radiation level at operator location behind the lead protective barrier was measured with and without flexi shield by simulating treatment conditions and dose was estimated based on the workload provided by the hospital.

Results: All electrical, mechanical, dosimetric and radiation safety parameters are within the specified tolerance limits. In-built safety switches, displays and interlocks were provided and functional. Measured radiation level behind the protective barrier was 0.9 mR/hr and 0.3 mR/hr without and with flexi shield respectively. Based on the work load provided by the institute as 30 patient/wk and average 30 min/patient assuming all intracavitary cases having maximum treatment time, the radiation level are found within the permissible dose limit of radiation worker.

Discussion and Conclusion: Performance of the Axxent EB system model 110 was found satisfactory. As the equipment is operated from inside the treatment room, operational safety plays a major role in radiation protection. Hence, a standard operating procedure (SOP) should be made and implemented by the user institute. The staff who are not familiar with radiation safety such as anesthetist, surgeon may be involved during procedure, hence it is the responsibility of the licensee to establish SOP such that non-essential staff do not remain in the treatment room when the beam is switched ON. Access control, warning signages should be used for this purpose. Staff should use the protective accessories such as flexi shields, lead protective barrier, lead aprons and follow the SOP. *Authors are thankful to the Medical Physicists of Hyderabad Institute of Oncology, Hyderabad and M/s. Rosalina Instruments, Mumbai for their support in this study.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-28: DOSIMETRIC COMPARISON BETWEEN TG-43 AND TG-186 IN LIP AND BUCCAL MUCOSA BRACHYTHERAPY IMPLANTS

Kamalnayan Chauhan 1, Priyanka Agarwal 1, Sarbani Laskar 1, Jamema Swamidas 1

Introduction and Purpose: TG-43 dose calculation formalism for photon emitting radionuclide sources used in brachytherapy, is based on the parameterization and superposition of single source dose distributions, obtained in liquid water under fixed geometry. Although, it is easy, fast and practically applied in clinics, it inherently lacks considerations for tissue and applicator heterogeneities, differences between absorbed dose in water and tissues, inter-seed attenuation, finite patient dimensions and dose contributions from electrons, can lead to under or overestimation of dose calculation. Model-based dose calculation algorithms (MBDCAs) have recently been emerged as a potential formalism for dose calculation in brachytherapy which involves tissue-inhomogenities and lack of scatter. It offers the possibility of departing from water-only geometries by modelling radiation transport in non-water media (tissues, applicators, air-tissue interfaces), resulting in a much more accurate dose distribution delivered to the patient. The present work provides the comparison between the dose calculation between TG-43 and TG-186 formalism for lip cancer and buccal mucosa brachytherapy Implant.

Materials and Methods: Five lip cancer and three buccal mucosa Brachytherapy implant patients previously treated in our hospital using TG-43 formalism (Oncentra Brachytherapy TPS v 4.5.3) were taken for this retrospective analysis. The mean(sd) of catheters in lip and buccal mucosa implants were 7 (2), and the number of implant planes were 2. CT images were transferred to TPS, followed by catheter reconstruction. The source activation was based on the clinical examination/or using the implanted markers. Dose prescription was done on basal points followed by geometric optimization on volume. Plan evaluation was carried out jointly by the treating physician and the physicist. Plans were made using TG 186 algorithm (Oncentra Brachytherapy TPS v4.5.3), keeping all the other parameters constant. A total of 16 plans were made. For this dosimetric analysis, a single brachytherapy fraction was considered. Dose prescription was 4Gy per fraction. The isodose volume covering 240%, 200%, 150%, 120%, 100% and 85% and 50% of prescription isodose were evaluated for TG 43 and TG 186 plans. The difference between these volumes in absolute difference was evaluated. In addition, Dose Homogeneity Index (DHI) was calculated and the difference between these plans were compared.

Results: The mean(±sd) of absolute differences of various isodose volumes was found to be 0.6(±0.2) cc and 1.0(±0.5) cc for lip and buccal mucosa implant respectively. Interms of percentage variation, the mean(±sd) difference was found to be 49(±21) and 30(±23)% for lip and buccal mucosa implant respectively. It was observed that TG 43 overestimates the dose more in higher isodose volumes as compare to lower isodose volumes. TG 186 plans were more homogeneous as compared to TG 43. The mean(sd) DHI of TG 43 vs TG 186 was 0.66(0.06) vs 0.76(0.08) and 0.73(0.02) vs 0.76(0.01) for lip and buccal mucosa respectively.

Conclusion: The difference between TG 43 and TG 186 algorithm was dosimetrically evaluated for brachytherapy implants such as lip and buccal mucosa. The difference between these two algorithms for the evaluated implants was found to be subtle. The clinical significance of these differences is not yet known.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-29: RECTAL AND BLADDER DOSE MEASUREMENTS IN THE INTRACAVITARY APPLICATIONS OF CERVICAL CANCER TREATMENT WITH HDR AFTERLOADING SYSTEM: COMPARISON OF TPS DATA WITH MOSFET DETECTOR

Navin Singh 1, Ashutosh Sinha 1, Shraddha Shrivastava 1, R Shanta Laxmi 1, Akanksha 1, L Satish Kumar 1, Teerth Raj Verma 1

Introduction: Brachytherapy of cervix carcinoma often results in high doses to surrounding structures, such as rectum and bladder. Therefore, these organs should be closely monitored.

Purpose: The purpose of this study is to evaluate the role of in vivo dosimetry with mobile MOSFET detector in HDR brachytherapy and to compare the actual doses delivered to organs at risk (as measured using in vivo dosimetry) with those calculated during treatment planning.

Materials and Methods: A total of 50 patients were treated with Microselectron HDR. Out of 50 patients, 26 patients were treated with the prescription dose of 7 Gy and the other 24 patients with prescription dose of 9 Gy. The Oncentra TPS version-4.5.2 was used for the brachytherapy planning and the evaluation of dose to the bladder & rectum. Further for in vivo dosimetry, portable MOSFET dosimetric system with two MOSFETs (TN – 502RD-H, SN: 33545 & 33546) and an electrometer (Team Best, Canada) were used for the measurements of bladder and rectum dose. A Siemens Somatom Emotion 16 slice CT simulator was used for CT images. For calibration of MOSFETs detectors, an acrylic cylindrical phantom with a center at hole and four holes in perimeter at 00, 900, 1800, and 2700 was scanned for 5 mm slices with MOSFET dosimeters and the data was exported to TPS. The brachytherapy plans were generated in which Ir-192 source was placed at a single dwell position in the center hole and prescribed dose as 50 cGy, 100 cGy, 200 cGy, 300cGy, 500cGy, 700cGy and 900 cGy at MOSFETs predetermined positions respectively. The calibration graph has been obtained to the dose up to 500 cGy, and the dose-response curve was found to be linear. MOSFET dosimeters in standard bias were kept in perimeter holes and recorded the dose online.

Results: The percentage deviation between the TPS-calculated maximum dose and the dose measured by MOSFET was found to be less than 5% in 31 patients, 5–10% in 8 patients, and 10–15% in 7 patients for rectum while dose deviation was greater than 15% in 4 patients where as in the bladder this percentage deviation was found to be less than 5% in 28 patients, 5–10% in 14 patients, and 10–15% in 4 patients. The dose deviation was greater than 15% in 4 patients for bladder.

Discussion: Our results showed that out of the total number of 50 insertions, the maximum rectum dose in 11 insertions (72.5% of treatments sessions) and the maximum bladder dose in 15 insertions (45% of treatments sessions) were higher than the 80% of prescribed dose to the point of dose prescription. This may be related to either the experience of radiation oncologist or to Oncentra treatment planning system. A more detailed study in this field is required to illuminate the cause.

Conclusions: The main reason for the differences between the measured and calculated doses was patient movement. To reduce the risk of large errors in the dose delivered, in vivo dosimetry should be performed in addition to treatment planning system computations.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-30: MATHEMATICAL APPROACH IN DETERMINING USE FACTOR FOR EQUIPMENT WITH ROTATIONAL DOSE DELIVERY TECHNIQUE

Amanjot Kaur 1, P N Pawaskar 1, G Sahani 1

Introduction: The optimizing parameters required for shielding thickness calculations of a radiotherapy bunker need to be appropriately estimated to arrive optimized protective barrier thicknesses. Use factor as one of the optimizing parameters is highly dependent on the treatment delivery technique of radiotherapy equipment. Therefore, it needs to be derived and calculated for the equipment by considering dose delivery technique available in the equipment. Hence values of use factors for Cyberknife, Tomotherapy, Halcyon etc. are different than that of Medical Accelerator having same photon energy. In this paper, use factor is mathematically derived for equipment in which source is continuously rotating around axis of rotation such as Helical Tomotherapy and Halcyon. Therefore, this formula is useful to calculate use factor to be used in arriving primary protective barrier thickness of bunker housing radiotherapy equipments using continuous source rotation around axis of rotation for dose delivery.

Purpose: Aim of this paper is to establish mathematical expression for calculating use factor of radiotherapy equipment wherein source continuously rotates around the patient for dose delivery.

Materials and Methods: For the equipment in which source rotates in a circular path around isocentre in vertical plane, an arbitrary point of interest taken on primary barrier located at a distance ‘d’ from isocentre is continuously exposed by the projected radiation field length till source moves from one position to another such that complete radiation field length is swept away through this point of interest. Further, it is also found that point of interest on the primary barrier located nearer to source still keeps on receiving primary radiation whereas at farther distance from source irradiation is already over for the same geometrical set up. Hence, angle of rotation at isocentre corresponding to the source movement from initial to final position (α) is found to be varying with distance of the location of primary barrier from the source (d). The beam opening angle at source for corresponding field length is mentioned as θ. The relationship derived in terms of α, θ, d and distance of source from isocentre (R) can be derived and expressed as follows: As source is continuously rotating around patient, based on definition of use factor (U), it can be expressed as follows:

Results and Discussion: The mathematical expression as derived for the use factor (U) as given in equation (1) shows that value of use factor varies as per the location of placement of primary barrier from isocentre and equipment design specifications such as field length and source to isocentre distance. Further, it is also observed that value of use factor increases when primary barrier is placed nearer to source whereas decreases when primary barrier placed farther away from the source for the same geometrical setup.

Conclusion: Using above mathematical expression, the use factor can be calculated for given θ, R and d. This study is helpful for those users who wish to design bunker for the machines that use rotating source geometry (e. g. Tomotherapy, Halcyon) and desirous to replace their existing teletherapy equipment with such equipment. Results obtained using above expression are found to be in good agreement with the studies of other authors Robinson et al. and Chuan Wu et al. for tomotherapy.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-31: RADIATION SHIELDING CONSIDERATION FOR HALCYON VAULT DESIGN

G Sahani 1, P K Dixit 1, P K Dash Sharma 1

Introduction: Halcyon is a new model of medical linear accelerator introduced by M/s. Varian Medical Systems, USA recently. Halcyon uses single 6 MV unflat beam (flattening filter free) with maximum definable field size of 28cm × 28cm at isocentre (SAD=100 cm). In order to reduce shielding requirements, manufacturer has introduced beam stopper having transmission of 0.1%. Radiation dose to the patient can be delivered either using volumetric modulated arc therapy (Rapid Arc) or static gantry through intensity modulated radiation therapy (IMRT) technique. The optimizing parameters required for shielding thickness calculations of a radiotherapy bunker need to be appropriately estimated to arrive optimized protective barrier thicknesses. The optimizing parameters are highly dependent on the dose delivery technique of radiotherapy equipment and hence needs to be estimated/calculated considering these aspects. Hence values of optimizing parameters and shielding requirement for the given design goal for Cyberknife, Tomotherapy, Halcyon etc. are required to be different compared to that of standard Medical Accelerator even for the same photon energy. Recently, one of the institutions approached Atomic Energy Regulatory Board (AERB) for obtaining layout approval for installing Halcyon from radiological safety view point. In this paper, authors have attempted to determine the appropriate values of optimizing parameters for calculation of radiation shielding thicknesses of walls/ceiling of the room for installing Halcyon.

Purpose: To find out adequate radiation shielding thicknesses of walls and ceiling of the Halcyon room so that radiation doses to the radiation workers and members of the general public shall not exceed their dose limits.

Materials and Methods: The minimum inner dimensions of room to house Halcyon unit, as recommended by the manufacturer, are around 5.9 m (length), 4.7 m (width) and 2.8 m (height). The minimum distance required between wall behind the gantry and isocentre of machine is 1.5 m. To estimate appropriate thickness of the protective barriers, the optimizing parameters such as workload, use factor and occupancy factor are established.

Workload:

  • i) For IMRT, Primary workload (Wpri) = 200cGy/pt x70 pt/dayx5 days/wk/PDD =1.11x105 cGy/wk at 1 m; and Head Leakage Workload (WL)=IMRT factor x Wpri = 4.44x105 cGy/min at 1m.

  • ii) For VMAT, Workload (WPri)=200cGy/pt x70 pt/dayx5 days/wk/PDD = 1.11×105 cGy/wk at 1 m.

Use Factor: Since the x-ray target is continuously rotating around the patient for treatment, Use factor (U) is determined to be 0.044 for VMAT. As product of workload and use factor for IMRT is found to be significantly higher than for VMAT, so the former is used for the calculations for conservatively safer. Using basic definitions and formulae provided in IAEA/NCRP reports, the radiation shielding calculations are carried out for primary and secondary barriers of all the walls and ceiling.

Results and Discussion: Calculated thicknesses of primary barrier of wall and ceiling are and found to be 135 cm and 150 cm of concrete (density 2.35 g/cc) respectively. Similarly, the thicknesses are calculated for the secondary wall (behind gantry), maze wall and secondary barrier in the ceiling and are found to be 140 cm, 80 cm and 140 cm of concrete(density 2.35 g/cc) respectively.

Conclusion: Based on the thicknesses calculated for all the walls and ceiling of Halcyon vault, a typical standard room layout drawing is proposed. From calculated thicknesses, it is inferred that any existing 6 MV accelerator or even telecobalt room can be used for installing Halcyon with minor modifications.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-32: A PRECLINICAL ANIMAL (RAT) MODEL OF RADIATION-INDUCED SKIN INJURIES TO STUDY VARIOUS WOUND THERAPY APPROACHES

Sukhvir Singh 1, Pradeep Goswami 1, Shweta Kulshreshtha 1, Aruna Kaushik 1, Mitra Basu 1, M K Semwal 1

Introduction: Current events throughout the world underscore the growing threat of different forms of terrorism, including radiological or nuclear attack. In the event of an attack, exposure of human tissue to radiation may result in both acute and chronic toxicities which can result in a range of symptoms and decreased quality of life. Skin is the primary organ of the victim which is directly exposed to the radiation through radioactive contamination, incorporation and energised particles of ionising radiation. Also, in radiotherapy skin can be a dose-limiting tissue for certain cancers, such as breast, head and neck. Cutaneous radiation damage to the skin is a major concern in radiation therapy of these cancers. Therefore, treatments or therapies are needed to protect and treat a wide range of acute and chronic radiation induced injuries to the human skin. Future therapies may revitalise the affected tissue and stop the further deterioration of the skin.

Purpose: The objective of the study is to provide a functional preclinical animal model of cutaneous radiation injury which can be used to evaluate various pharmaceutical formulations in protection, decontamination, decorporation and healing of radiation-induced skin injuries. A protocol was designed to apply cobalt-60 radiation to the skin of rats while sparing the body and internal organs.

Materials and Methods: A circular cone of 20 mm diameter opening was designed locally to be used as a tertiary collimator for Bhabhatron-II teletherapy unit. Output factor, off axis ratio and beam profiles were measured using ionisation chamber and Gaffchromic film as per the small field dosimetry protocol. In order to find appropriate dose for wound creation, each rat, with skin pulled away from the body was irradiated for 25.2 Gy, 34.6 Gy, 45 Gy, 55 Gy and 65 Gy in a single fraction. Following multi dose trial, five additional wounds were created using 45 Gy dose. The photographs of each wound were taken daily and followed till 65 days post irradiation. The images were analysed for percentage ulceration created at given point of time using Matlab image processing program.

Results and Discussion: The radiation induced wounds were formed in more than 60% of irradiated skin area within 12 days at all doses except at 25.2 Gy dose (52% wound at 16 days). By day five, all irradiated areas showed acute skin reactions like erythema, mild dryness, and hair loss. On the day ten, intense skin reactions started appearing and by the day 14, wet desquamation started appearing with higher doses (55 Gy and 65 Gy). The peak ulceration was ranged from 65.3% to 79% between 12 to 16 days postirradiation. The peak ulcerated area increased from 65.3% at 34.6 Gy to 75% at 45 Gy and then remained nearly 79% at 55 Gy and 65 Gy. The mean wound size calculated from day 10 to day 65, increased with the dose and showed a strong correlation (R=0.882) with the dose. In five additional rats irradiated with 45 Gy, the peak wound size ranged from 54.2% to 74.8% with a mean of 63.6% (SD = 3.1%) at 16th day postirradiation. The average wound size reduced to 23.1% (SD = 5.6%) at 65th day showed significant healing.

Conclusion: The preclinical animal model has been developed which is being used by various biological and pharmaceutical scientists to study biochemical mechanism of radiation induced wound healing and development of interventions for rapid healing of these wounds. This model may also be useful in the development of novel therapies to reduce skin reactions in radiother.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-33: DESIGNING AND STUDY ON RADIOLOGICAL PROPERTIES OF IN HOUSE DEVELOP LUNG PHANTOM

Priyusha Bagdare 1, Om Prakash Gurjar 1

Introduction: Phantoms are the devices which have been used in medical physics since the beginning. Soon after the discovery of X-rays physicist develop phantoms for diagnostic as well as for dosimetric purpose. Earlier it was thought that in order to quantify the dose delivered to a tissue of interest; the measurement should be made on the tissue itself. When the harmful effects of radiation were realized, the need for tissue substitute became clear, and thus the concept of phantom was born. As human body consists of water thus the dose measurement is done on the homogeneous phantom which is water equivalent. Whereas advance diagnostic techniques clearly suggest that the human body is complex heterogeneous medium comprises of varied densities. To study the radiological properties of photon inside heterogeneous medium thus necessitate developing a heterogeneous phantom for dosimetric purposes. However, the inconsistency of tissue equivalent materials still presents a large obstacle. The selection of the appropriate materials is critical to the design and function of any type of phantom. Though the anthromorphic phantoms are commercially available but they are not cost effective and as accurate as human tissues. Present study focuses on designing of in house cost effective lung phantom (slab-wooden dust-slab) SWS and to study its radiological properties for 15 MV photon.

Purpose: To study the designing of in house developed lung phantom and radiological properties of 15 mega voltage (MV) photon energy inside it.

Materials and Methods: The density of SP34 slab, wooden dust of pine, and thoracic region of 20 patients were calculated using computed tomography (CT) images. The depths of isodose curves of 100%, 95%, 90%, 85%, 80%, 75%, 65%, 60%, 55% and 50% were measured in CT images of both the mediums on TPS for 10×10 field size for 15 MV photon beam. Patient-specific quality assurance (QA) was implemented using homogeneous SP34 slab phantom and heterogeneous SWS phantom for 15 patients.

Results: The mean density of wooden dust, slabs, soft tissue, chest wall and lung was found to be 0.271, 0.994, 0.980, 0.947 and 0.287 gm/cc respectively. The isodose depth (of 100%, 95%, 90%, 85%, 80%, 75%, 70%, 65%, 60%, and 55%) for patient (2.83, 4.95, 7.03, 9.38, 11.04, 13.08, 15.14, 18.46, 20.29 and 20.89 cm) and for SWS (2.84, 5.02, 6.90, 9.07, 10.95, 12.60, 14.19, 17.75, 19.83 and 22.06 cm) are found to be approximately same. Furthermore, the mean percentage variation and standard deviation between the planned and measured doses in patient-specific QA was found to be 1.04, 0.64 and 1.41 and 0.80 respectively for SP34 and SWS phantom.

Discussion and Conclusion: Our aim was to design a heterogeneous phantom which can mimic the actual patient. The radiological properties of developed in house phantom are found to be equivalent to that of the actual thoracic region of human. Patient-specific QA was performed for 15 IMRT cancer patients. The variation between the measured dose and the planned dose calculated by TPS is found to be more in SWS phantom than the SP34 Phantom. Dose calculations were done by anisotropic analytical algorithm (AAA) which does not give accurate results when density variations are there. Thus we got more variation in SWS heterogeneous phantom as compared to homogenous SP34 phantom.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-34: AUTOMATION OF DELAY DECAY TANK FOR THE MANAGEMENT OF RADIOACTIVE WASTE

Kirti Tyagi 1, Partha B Mukherjee 1, Deboleena Mukherjee 1

Introduction: Department of Nuclear Medicine of INHS Asvini provides diagnostic and therapeutic facilities to its clientele using radioisotopes at this tertiary care hospital of Indian Navy. This department has been authorized by AERB for providing high dose radionuclide therapy to the patients. A nuclear medicine department which uses I-131 as a radioactive source for treatment of thyroid cancer, needs the patients to be admitted in wards for some days and their waste product is drained into separate waste management facility which is called delay tank. A significant amount of radioactive waste is generated from the patient. This radioactive waste is stored in delay decay tank for a minimum period of about 10 half-lives.

Purpose: In this paper we are describing the work taken up for renovating and upgrading the delay decay tanks from manual to automatic mode so that there will be no need of manual intervention in operating valves of the tanks while releasing the human waste containing decayed activity.

Materials and Methods: Liquid wastes generated from I-131 administrations are collected in the delay tank. In our Nuclear Medicine department there are two isolation rooms for administering high dose of radio iodine ablation therapy and sewage lines of these rooms are connected to the twin concrete tanks located in the area at ground level behind the department. Both the tanks are of a size of 4m × 2m × 1.5 m (external dimensions) each and can accommodate approximately 10,000l of effluents. Earlier the tank's valves were operated manually at the time of releasing the decayed activity. The tank lids were also needed to be lifted up for checking the levels within the tanks. The automation of storage and release system of tank has resolved these efforts. The automation of radioactive sewage tank system has been carried out to monitor the volume and levels of effluent sewage in the tank and releasing the same to main drain once safe level of radiations has reached. The electronic system comprises of:

  1. Programmable logic circuit (PLC) – to control motorized valves and to monitor fluid levels in both the tank. PLC are programmed to control motorized valves in automatic and manual mode.

  2. Control panels (02 in no.) – one control panel was installed near the delay tank with level display and push button control system for motorized valves. Second control panel installed with 12 inch LED touch screen panel in the department.

  3. Motorized valves-corrosion free, maintenance free valves of suitable size with manual and automatic controls.

  4. Pump system.

  5. Computer Software –to control automation of tanks.

Results and Discussion: The automation of radioactive sewage delay tanks at INHS Asvini is one of its kind, fully automatic and digitalized delay tanks in all armed forces hospitals. Automation has been executed and included renovation of existing delay tanks and installation of six motorized valves, ultrasonic sensors for monitoring effluent level, electronic relays, programmable logic circuit, touch screen control panel, push buttons for manual control and LAN port for desktop control.

Conclusion: Generation of radioactive waste shall be kept to the minimum. The automation of delay decay tank has reduced the dependency on the workers engaged in manual opening of lid and valves of the tank. The push button control panel has reduced the radiation exposure to the operators. The remote controlled valves helps in preventing any accidental overflow of the tanks by constantly monitoring the effluent's levels.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-35: PROTON RADIOGRAPHY STUDIES USING GEANT4 SIMULATIONS

B G Alaka 1, Deepak Samuel 1

Introduction: Proton therapy is the recent trend in cancer treatment. The depth-dose profile of proton exhibits a flat plateau followed by a sharp peak known as Bragg peak where the maximum dose is deposited. The proton range should be known accurately to deposit the maximum dose at the tumor site. Any uncertainty in proton range causes partial covering of the tumor or zero dose to the tumor which severely damages the surrounding healthier tissues. Presently proton ranges are calculated from X ray Computed Tomography (XCT) images in which X-ray Hounsfield Units (HU) are converted into water equivalent path length (WEPL) using a calibration curve. Any error in the calibration will translate into an error in the proton range. This can be reduced if protons are used for imaging instead of x-rays-a method commonly known as proton radiography recent technique called as Energy Resolved Dose Functions (ERDF) seems to be promising for clinical applications due to its simplicity. The relation between ERDF and WEPL are well established experimentally. In this study, the proton radiography setup is simulated and studied its efficacy in imaging a solid sphere.

Purpose: To study the image quality and WEPL accuracy of proton radiography using GEANT4 simulation.

Materials and Methods: The simulation environment was developed using Geant4 libraries. Proton beams varying from 50 MeV to 250 MeV are shot using a particle gun and a water wedge of dimensions 5 cm x 5 cm x 30 cm (HxWxD) is placed in that path of proton beams and the dose deposited at the exit of the wedge is measured using a simple detector geometry from which the dose function is estimated (ERDF). A solid (water) sphere of radius 10 cm is now kept in the path of the beam and the energy is scanned as before. By comparing the pixel wise ERDFs with the wedge ERDFs, we retrieve the WEPL image of the sphere.

Results: ERDF has a unique dose pattern as a function of energy for a particular WEPL. WEPL can be calculated by using the depth at 80% of the proximal fall off of the ERDF in water. ERDF for WEPL range of 0 cm to 30 cm are generated from wedge simulation data. By matching the ERDFs pattern from the sphere with the wedge, WEPL for sphere is estimated and compared with the calculated WEPL. The results show a good match between the WEPL from the simulation with the WEPL from the calculations.

Discussions and Conclusions: Simulation of single detector proton radiography yields convincing results in evaluating WEPL accuracy. Beam energy used in simulation need to be varied in smaller steps for further improvement of the accuracy of WEPL. Scattering effects are not included in the present study. Beam parameters like mean energy and energy spread need to be optimised for extending this method for clinical applications.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-36: INDIGENOUSLY DEVELOPED COST EFFECTIVE HETEROGENEOUS PEDIATRIC PHANTOM

C Senthamil Selvan 1, C S Sureka 1

Introduction: There is a spectacular technological development and advances in radiodiagnosis and radiotherapy while imaging or treating the cancer patients. The patient absorbed dose measurement is very essential to monitor dose level, but few verification methodologies are possible to know the radiation. A phantom study is a compatible and easy way to verify the dose level. Phantoms can be classified into two major categories such as computational phantoms and physical phantoms. Tissue equivalent material of polymethyl methacrylate (PMMA) is used to fabricate physical pediatric phantoms in order to evaluate the doses received by patients.

Purpose: To fabricate a heterogeneous pediatric phantom and measure organ dose in 1 year, 5 years and 10-year-old pediatric phantoms.

Materials and Methods: Different age group pediatric patient anatomical information has occurred from Computed Tomography (CT) scan. After the simplification of the CT images each slice and its corresponding data, including thickness and size were drawn using the AutoCAD software. Then the output of the AutoCAD images was transferred to a CNC machine. The 10 mm thickness of PMMA sheets with a density of 1.17 g/cm3 was cut into many pieces of slabs, the dimension was 26×20 cm2 for 10-year baby, 24×18 cm2 for 5-year baby and 20×14 cm2 for 1-year baby. In each slab drilled two 10 mm diameter hole was made which this hole is fixed in CNC machine to avoid motion. We made provision of OSLD nanoDots cavity in critical organ region at the phantom. The phantom irradiations were performed in digital radiography RADSPEED 80 equipment (Shimadzu Medical India Pvt., Ltd.,). Sets of nanoDot were positioned in the 5-year-old phantom at different locations like Head, Neck, Lung, and Abdomen. The operating parameters were selected 60kVp to 120kVp with an interval of 20 kVp at 50mAs, 100mAs and, 150mAs. The field size was fully opened and the phantom placed at 70cm away from the focal spot.

Results: The measured and calculated average absorbed dose is 0.35mGy for 60kVp and 50mAs, 0.40mGy for 80mGy and 50mAs, 0.48mGy for 100kVp and 50mAs, 0.56mGy for 120Vp and 50mAs, 0.56mGy for 60kVp and 100mAs, 0.61mGy for 80kVp and 100mAs, 0.72mGy for 100kVp and 100mAs, 0.92mGy for 120kVp and 100mAs, 0.72mGy for 60kVp and 150mAs, 0.82mGy for 80kVp and 150mAs, 0.95mGy for 100kVp and 150mAs, 1.13mGy for 120kVp and 150mAs respectively.

Discussion: The characteristics of the developed heterogeneous pediatric phantom were ensured and optimized to measure the radiation exposure level in diagnostic radiology. From this study, it is inferred that the radiation dose in pediatric phantom was varying with respect to tube voltage and tube current. The observation of organ absorbed dose in those phantoms was slightly varied due to a different location at the phantom. The dose received by the lung and a head region is more when compared to other parts of the body. Based on these, the present would be used to measure organ dose in the pediatric patient.

Conclusion: A unique methodology has been developed to construct heterogeneous pediatric phantoms for dosimetry studies. The total cost of these phantoms (3 Nos.) is less than Rs. 3 Lakh. The cost-effective phantom is well suitable for use in radio diagnostic procedures, to measure the actual dose delivered to critical organs while performing imaging.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-37: MACHINE LEARNING-ENHANCING PRODUCTIVITY IN NETWORKED ONCOLOGY CENTRES

A Pichandi 1

Purpose/Objective(s): The field of Radiation oncology has grown leaps and bounds in the past 3 decades. The technology has revolutionized the treatment outcomes. In the recent past number of cancer care centers in India has grown steadily. This lead in to the networking of these centers to enhance the productivity, quality of care in these networked treatment centers. The knowledge of machine learning, deep learning and artificial intelligence have played a significant role in this regard. Machine Learning is study of algorithms that improve their performance at some task with experience. We have used RapidPlan (Varian Medical Systems, Palo Alto, CA) for planning. RapidPlan is a machine-learning tool that studies best practices from past successful treatment plans and creates knowledge-based treatment models that are applied to improve the treatment plans for future patients. These RapidPlan models help to quickly generate and validate new high-quality treatment plans based on shared expertise among centres. We have taken 3 important sites for this purpose, i. e., Head & Neck, Pelvis and Craniospinal irradiation (CSI). The Head and neck and Pelvis under validation. The CSI clinical plans were generated and compared with Rapidplan done in one our unit on various treatment parameters. CSI remains technically demanding, with potential for treatment filed overlap or gaps to yield unacceptable dosimetric target heterogeneity and time consuming for planning. RapidPlan model for CSI were created using dosimetric inputs from Helical Tomotherapy (HT) planning to guide optimization and to generate plans with Volumetric Modulated Arc Therapy (VMAT) technique in a different planning system and then to test the performance of this model against user plans.

Materials and Methods: Thirty six CSI patients treated with HT were chosen to create Rapidplan model. All image set had organ at risks (OAR) such as Heart, bilateral Kidneys, Lungs, Eyes, Rectum, Bladder contoured and deliverable plans were created using Tomotherapy Volo planning system. The Radiotherapy dose dicom file were then transferred to Varian Eclipse planning system (Version 13.7) to train KBTP models. For the purpose of training KBTP model, we considered five OARs say Heart, bilateral Kidneys, Lungs to extract dosimetric data. The model quality was evaluated checking the model goodness of fit statistics for each structure, with the coefficient of determination R 2 (between 0 and 1: the larger, the better) and the average Pearson's chi square χ2. The published model was then used to guide VMAT Optimization and to generate deliverable plans on an independent set of 20 adult patients using three isocentre. Resultant model plans were then compared and analyzed against user plans using VMAT and HT. The planning target volume (PTV) coverage were compared using the Uniformity Index (UI). Four dose-volume points such as V5, V10, V20 and mean doses were used to compare the OARs doses.

Results: Out of 36 patient dosimetric data extracted, 27 dataset were used to train Rapidplan model excluding potential poor outliers from data set. R-square values of trained model as goodness of fit for OARs were Heart 0.623, Left Kidney 0.761, Right Kidney 0.887, Left Lung 0.965 and Right Lung 0.903. UI for PTV were 1.03 ± 0.02 and 1.06 ± 0.01 for KBTP model plan and user VMAT plan respectively. Mean dose to Heart, Right and left Lungs, Right and left kidneys were 400 cGy ± 59, 462.1cGy ± 53, 406.8 cGy ± 40, 431.7cGy ± 30, 406.6 cGy ± 31 for KBTP plans and 383.2 cGy ± 29, 423.8 cGy ± 17, 393.1 cGy ± 44, 557.6 cGy ± 20, 557.2 cGy ± 19 for user plans respectively. There is no significant improvement in OAR dose volume values between plans. The targets heterogeneity and global maximum dose were found to be high in VMAT user's plan by 2-3 % and closely fit to HT user's plans when compared with model plans.

Conclusion: RapidPlan CSI models are suitable for generating clinically acceptable VMAT plans and planning time is greatly reduced. An adequate choice of the objectives in the model is necessary for the trade-offs strategies. A possible clinical benefit as reduced toxicity remains to be tested.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-38: DOSIMETRIC VALIDATION OF ACUROS XB PHOTON DOSE CALCULATION ALGORITHM ON AN INDIGENOUSLY FABRICATED LOW DENSITY HETEROGENEITY PHANTOM

Girigesh Yadav 1, Manindra Bhushan 1, Lalit Kumar 1, Kothanda Raman 1, Suhail 1, Gourav Kumar 1, Praveen Ahlawat 1, Munish Gairola 1

Introduction: Precise confirmation of accuracy in dose calculation is an important aspect of quality assurance procedure in radiotherapy. Slab phantom of uniform density (water equivalent i. e 1g/cc) used for patient specific QA in most of radiotherapy department of developing countries. This is due to high cost of commercially available heterogeneous phantoms and limited access to advance technologies. Owing to known effect of tissues heterogeneity on dose distribution pattern and dose calculation, there is a need to fabricate a cost effective low density heterogeneous phantom. This can truly represent the accurate dose calculation in low density heterogeneous medium.

Purpose: The aim of the study is to fabricate a cost effective low density heterogeneous phantom using the combination of PMMA and racemosa wood. This study also validates the Acuros XB algorithm in low density heterogeneous medium of this fabricated phantom.

Materials and Methods: This study measured the Hounsfield units (HU), relative electron density, mass density for racemosa wood and PMMA. Point dose measurement were done at 6.5, 11 and 17.5 cm depths in heterogeneous PMMA-Racemosa-PMMA (PRP) and homogenous PMMA mediums and compared against AAA and AXB calculations. This experiments was performed using a 6 MV photon beam and dose calculations were performed on eclipse TPS. The dose was reported in dose-to-medium and dose-to-water mode for AXB and AAA algorithms respectively. The grid resolution of 2.5 x 2.5 x 2.5 mm3 was used for dose calculations.

Results and Discussion: The measured HU, relative electron density, mass density were-726.5, 0.273g/cc, 0.212g/cc and 201.8, 1.201g/cc, 1.175g/cc for racemosa and PMMA respectively.

For AAA algorithm, the mean dose variations from IC measurement were 0.73%,-0.99% and-1.39% at depth of 6.5 cm, 11 cm and 17.5cm respectively, in homogeneous medium of PMMA. Similarly, mean dose variations from IC measurement were-1.67%,-1.45% and-1.50% at depth of 6.5 cm, 11 cm and 17.5cm respectively, in low density heterogeneous medium of PRP phantom.

For AXB algorithm, the mean dose variations from IC measurement were-1.01%,-2.41% and-1.85 % at depth of 6.5 cm, 11 cm and 17.5cm respectively, in homogeneous medium of PMMA. Similarly, mean dose variations from IC measurement were 1.50%, 1.66% and 1.81 % at depth of 6.5 cm, 11 cm and 17.5cm respectively, in low density heterogeneous medium of PRP phantom.

The percentage dose variation between AAA and AXB were also calculated and compared in homogenous medium of PMMA and low density heterogeneous medium of PRP phantom. The variation of-2.58%,-1.44% and-0.47 % were found at depth of 6.5 cm, 11 cm and 17.5cm respectively, in homogeneous medium of PMMA. The variation of-3.13%,-3.06% and-3.26 % were at depth of 6.5 cm, 11 cm and 17.5cm respectively, in heterogeneous medium of PRP phantom. This dose difference in AAA and AXB algorithm exist due to the difference in dose calculation methods employed in both algorithms. The AXB calculations are sensitive to medium composition and characterization as they calculate radiation transport in the medium. In contrast, the AAA calculations model the medium was water of different densities.

Conclusion: Based on the physical and radiological properties studied for racemosa, it can be concluded that racemosa can simulate lung region of a human body. It can be use to fabricate a cost effective low density heterogeneous phantom in combination with PMMA and other suitable materials. The experimental validation of AXB calculations also concludes that AXB provides comparable results to AAA, in low density heterogeneous medium.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-39: IMAGE BASED DOSIMETRY QA TOOLS FOR ADVANCED RADIOTHERAPY

Rajesh Kumar 1, Ankit Srivastava 1, R K Chaudhary 1, N R Kakade 1, S D Sharma 1, D Datta 1

Introduction: Patient-specific pre-treatment dose verification for safe and effective implementation of advanced radiotherapy techniques such as IMRT/VMAT is strongly recommended for all the patients. A number of commercial systems are available for pre-treatment dose verification in IMRT/VMAT; however these systems are costly, having limited features and dedicated to a particular dosimetry system. In view of this, an indigenous, versatile image based dosimetry QA system was developed. This paper describes features of indigenously developed image based dosimetry QA system for advanced radiotherapy techniques.

Materials and Method: Image based dosimetry QA tools for advanced radiotherapy has been developed using Visual Studio C++ programming language. It has 6 core menu items and there after a number of sub-menu for different activities/task. Applications of this tool open with four windows namely 1, 2, 3 and 4. Window 1 and 2 are used to import the images for analysis while Window 3 and 4 are used to display their analysis/ computed results. The system have feature for 2D and 3D analysis of data set. It can be used for relative as well as absolute dose analysis. As is the case with other QA systems, the study in this tool start with calibration. Pre-digitized calibration image (in tiff format) can be loaded or scanner can be called to digitize the calibration image. The calibration function (where A, B and C are constants) was used to generate the calibration curve. These constants were determined by solving equations with data of different calibration doses. In the sub-menu called BUILD, digitized image is converted into dose-map, isocentre and ROI are fixed, and the dose-map is converted into DICOM RT having tag of the patient details as communicated by the TPS. This QA system also has a dedicated database where all the data including the TPS calculated dose data as well as measured dose data are stored. Using the sub-menu called LOAD, these data can be loaded into windows 1 and 2. This system is equipped with a number of analysis toolbar such as gamma, DTA, dose difference, isodose comparison, profile comparison, histogram analysis, normalization option, data manipulation with simple arithmetic operation, data movement. Report of the analysis is displayed by the system. Physics QA for analysis of field width, flatness, symmetry, penumbra for flattening filter as well as flattening filter free beam also have been added. The analysis of these parameters are done using international/AERB protocols. To demonstrate its performance, pre-treatment dose verification for an IMRT case of H&N cancer was carried out. QA plan was generated using Eclipse Treatment planning system. The TrueBeam medical linear accelerator having 6 MV flattened filter beam was used for this study. The Gafchromic film strips were irradiated in the range of doses 60, 140, 300 and 340 cGy for the purpose of calibration.

Results and Discussion: The calibration curve was generated. The dose-map using the generated calibration curve was computed. Computed dose-map was aligned, isocentre and ROI was fixed. The computed dose-map was compared with TPS generated dose map. Analysis of results using gamma reveal that 99.19% point passed the set criteria of 3% and 3 mm. Analysis of results also reveal that the isodose lines and dose profiles of both data sets are well correlated with each other. Further, result of dose difference relative to the local and global dose (dose at isocentre) indicates that differences are mostly around 1% and -1% respectively. It should be noted that dose difference are found to be significantly higher than typical accepted dose difference of 3% but the frequencies are negligible.

Conclusion: The system's performance has been thoroughly evaluated and found satisfactory in comparison to the similar commercial system.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-40: HYBRID TECHNIQUE FOR BREAST CANCER: A DOSIMETRIC COMPARISON OF HYBRID VMAT AND HYBRID IMRT PLANS

Karunakaran Balaji 1, Balaji Subramanian 1, T Moorthi 1, K Sathiya 1, C Anu Radha 1, V Ramasubramanian 1

Introduction: Hybrid planning technique is an innovative method which combines conventional 3DCRT and VMAT/IMRT plans with different dose proportions in a single fraction. In published studies, the average proposed dose ratios for 3DCRT and VMAT/IMRT are 75% and 25% respectively. Published Studies preferred IMRT for whole breast irradiation than VMAT. Further, there is no study on the feasibility of flattening filter free (FFF) photon beam in 3DCRT based hybrid techniques.

Objective: In the present study, hybrid IMRT and hybrid VMAT plans for whole breast irradiation were compared and the feasibility of FFF beam in hybrid plan was studied.

Materials and Methods: CT scan data of 10 early stage left-sided breast cancer patients were used for this study. The whole breast PTV was prescribed to 50Gy (2Gy/fraction) using 6X and 6X-FFF photon beams of Truebeam STx accelerator. The 3DCRT plan was prescribed with 75% dose per fraction and used 2 tangential fields with 2 cm air margin to accommodate breathing movement and setup variability of breast PTV. The VMAT and IMRT plans were prescribed with remaining 25% dose and created by keeping the 3DCRT plan as a base dose plan during optimization. All plans were normalized to target mean and statistically compared using various PTV dosimetric variables and organs at risk (OAR) doses.

Results and Discussions: Both H-VMAT & H-IMRT plans showed comparable PTV coverage, conformity, uniformity and gradient indices. Many clinical studies have warned about the potential risk for lung and heart even at low doses. The ipsilateral lung mean, V5Gy were significantly less in H-IMRT (p< 0.01) whereas V20Gy, V40Gy were comparable. The heart mean, V5Gy, V40Gy were significantly less in H-IMRT (p< 0.01) whereas V25Gy was comparable. The contralateral lung, contralateral breast and normal tissue mean doses were significantly less in H-IMRT (p< 0.04). The MU and treatment time were significantly less in H-VMAT (p< 0.0001). The hybrid plans with FFF beam showed statistically similar results compared to those plans with flat beam.

Conclusions: The H-IMRT plan achieved significantly better sparing of OAR doses compared to H-VMAT. Less low dose bath to the heart, lung and contralateral breast should be the goal since the breast patients are long term survivors. The use of FFF photon beam is feasible in hybrid plans with 3DCRT as one of component for whole breast irradiation.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-41: APPLICATION OF IAEA-AAPM TRS 483 FOR SMALL FIELD DOSIMETRY WITH TOMOTHERAPY: AN INITIAL EXPERIENCE

Suryakant Kaushik 1, Vinay Saini 1, Sudarshan Kadam 1, Chandrashekhar Tambe 1, A Sutar 1, Rajesh A Kinhikar 1, Deepak Deshpande 1

Introduction: Recently IAEA-AAPM has published TRS 483 on small field dosimetry. It guides in detail about the testing the code of practice for small field dosimetry. Nowadays, advanced treatments are delivered with either static intensity modulated radiotherapy (IMRT) fields or rotational IMRT fields by utilizing small beamlets. The appropriate use of the detector with minimal uncertainties is the need of the time. This study deals with relative dosimetry of the small fields that can be generated using the MLCs of Tomotherapy Machine by using various detectors.

Purpose: To determine the output factors (O. F) and TPR (20, 10) for a range of field sizes in water and virtual water with three different detectors on Tomotherapy Hi-Arts machine.

Materials and Methods: Tomotherapy Hi-Art System (Accuray, USA) having 6MV FFF energy was used for the dosimetry. Three different detectors (PTW Pin Point, IBA CC01, IBA EFD 3G Diode) with Tomotherapy Electrometer (8-channel) System were used. Tomotherapy water tank was used for TSD setup measurements in water whereas Scanditronix Wellhofer Gmbh I’mRT verification Tool Phantom with some customization was used for TSD and TAD setups in virtual water phantom. The Pinpoint and CC01 chamber axis was placed perpendicular to the beam axis, whereas the axis of EFD diode was placed parallel to beam axis. The detector positioning was crucial but was verified with the MVCT imaging for every measurements. Range of field size used varied from 5cm X 40cm to 1cm X 0.625cm.

Recommendations from TRS 483 were followed for O. F measurements. All O. F were normalized for the reference field size 5cm X 10cm. All measurements were carried out at 10 cm depth, and the O. F measured with three detectors were compared. In addition, the uncertainty budget was also estimated and reported.

Results: EFD diode showed the highest uncorrected O. F for the smallest field in both water TSD setup and virtual water TAD setup. Largest variation among detectors for uncorrected O. F were 10.73%, 5.38% and 14.69% for water TSD setup, virtual water TSD & TAD setup respectively. The corrected O. F of EFD and CC01 were comparable up to 3 decimal points in water. The corrected O. F of EFD were higher in comparison to CC01 with a difference of 1.7% and 11.71% for Virtual water TSD and TAD setup respectively. Both uncorrected and corrected O. F for detectors are in good agreement within 3% except field sizes smaller than 2.5cm. Maximum variation of TPR (20, 10) for reference and smallest field size among detectors was 0.34% and 1.5% respectively. The maximum total uncertainty (Type A: meter reading; Type B: time, couch position, Electrometer, Correction factors) of 3 detectors were found at smallest field sizes which are 1.25%, 3.55% and 3.64% for EFD, CC01 and Pinpoint respectively in water SSD technique.

Discussion: The output correction factor for PTW Pin Point was available upto equivalent field size 2.5 cm, so corrected O. F was not calculated for F. S <2.5cm. Due to the constraint of bore size, Tomo water tank cannot be used for TAD and TPR (20, 10) measurement in water.

Conclusion: The relative O. F for three detectors on Tomotherapy machine for 6MV energy were investigated for small fields. At the smallest field size, EFD showed the highest corrected O. F among the three detectors concluding its best suitability at small fields. The corrected O. F for the various detectors in water were found to be in close agreement (<3%) for all field sizes whereas, in virtual water the corrected O. F showed some variation for smaller field sizes (<2.5cm). The TPR (20, 10) values obtained for various detectors showed a similar trend, which increases with increase in equivalent field size.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-42: DOSIMETRY CHALLENGES ON VARIAN HALCYON LINAC–SIMPLIIFED WITH AUTOSETUP OF PTW BEAM SCAN

P Nagendran 1, B Arunshiva 1

Introduction: Varian Halcyon Linear Accelerator is the latest and advanced, fast treatment machine with 100% Image Guided treatment for all patients. The linear accelerator is designed like a Ring and is delivered pre-commissioned on the Varian Eclipse Planning System. Beam data measurements are required to ensure and verify the quality of the pre-commissioning data but these measurements are challenging for a physicist because there is no field light, no ODI and no laser on the Isocentre plane. The newly launched RFA, PTW BEAMSCAN, water tank was used for the first time in India to commission South East Asia's First Halcyon unit at Sterling Hospitals, Rajkot. The RFA has to be put on the couch top for measurements, so the couch tilt, tank level, water level, isocentre position is very difficult to align correctly. The BeamScan RFA was used using the Autosetup for the Acceptance Testing and Commissioning of the Halcyon Linear Accelerator.

Materials and Methods: The bore of the Halcyon Linac is 1 metre diameter and hence the ideal Beam Scan tank dimension to fit perfectly in to Halcyon bore is 70 cm width x 70 cm height, minimum. The weight of the water filled tank creates a Couch of sag of 3 mm to 5mm depending on the water level. PTW BeamScan has Auto setup feature and hence no need to do any manual adjustment and if couch sags due to more weight, auto setup mode will adjust as per the water level. The water tank is placed on the couch, aligned with the bore laser and the device homing is done using the wireless device. Water is filled upto the laser level for SSD 100cm and water surface is levelled is asserted using MV imager at gantry 90°. To setup the tank for SSD 90cm, additional 10 cm water is filled above laser level and the imaging is done at gantry 84.3°. The MV imager was also used to check the position of the reference detector using a brass build-up cap. The detector position, tank levelling and beam centering including detector at water surface is done with tank auto-levelling process. Once all setup was completed, PDD, In-plane and Cross-plane profiles for the filed size from 2X2cm to 28X28cm in the depths 13mm, 50mm, 100mm, 20mm were measured. The same setup was used for the absolute dose measurement using 0.6cc chamber for the output calibration and also for the TPR20/10 measurement.

Discussion: The Haclyon machine is a novel machine with pre-loaded beam data on the Eclipse Planning System. Hence, the validation of the beam data is very much necessary to gain confidence on the linear accelerator. At Sterling Cancer Hospitals, Rajkot the validation was done using PTW BeamScan RFA. The profiles were overlapped with the pre-loaded data in the Eclipse Planning System and found to be a perfect match. Automatic function of BeamScan constitutes as perfect solution for measurement. The big advantage of the BeamScan is reduced setup time with wireless auto setup with a maximum of 25min. Consistency of PDD at 10cm depth, fs= 6X6, was CoV=0.24, and Consistency of Inplane at d=10cm, fs=6X6, difference is-0.16%,-0.20%,-0.14%,-0.24% and Crossplane difference 0.02%,-0.12%, 0.11%, 0.08%.

Conclusion: Automatic Setup of RFA is very much needful for Commissioning a machine like Halcyon Linear Accelerator. PTW BeamScan is a perfect solution for the same. The consistency, accuracy and precision of the Halcyon machine is assured with consistency of measurements done with the BeamScan RFA.

J Med Phys. 2018 Nov;43(Suppl 1):S14–S38.

OP-43: INVESTIGATION OF THYROID RADIATION DOSES DURING MAMMOGRAPHY

Hema Joshi 1, Rajni Verma 1, Arun Chougule 1

Purpose: The use of mammography as a diagnosis has increased in recent years, being currently, the most sensitive technique and mostly performed for early diagnosis of breast cancer. Consequently, there is an increase in the number of exposures to radiation on breast and adjacent organs, appearing the need to evaluate the dose absorbed by risk organs, like thyroid gland. This study was planned to quantify the scatter radiation dose to right and left lobe of thyroid gland during routine screening mammography examination.

Methods: The radiation dose to the skin overlying the thyroid was measured for 100 women undergoing routine mammographic screening within the age group of 35-60 years and BMI range of 20-30 screened on Hologic M-IVTM screen-film mammography system single handedly. Measurements were made using optically stimulated luminescent dosimeter (OSLD) detectors taped appropriately to the skin overlying on right and left lobe of thyroid gland. The radiographic parameter ranges used was KV (28-30), mAs (10-24). The standard technique of two craniocaudal views (CC) and two mediolateral oblique views (MLO) was used during screening examination.

Results: An average skin thyroid dose of 0.40 +/-0.26mGy per mammographic examination was measured with measurements ranging from 0.10 to 0.79 mGy.

Conclusions: The single examination doses were not found significant but higher frequency of screening examination will increase the risk of radiation doses.


Articles from Journal of Medical Physics are provided here courtesy of Wolters Kluwer -- Medknow Publications

RESOURCES