09:00
Added to your list of favorites
Deleted from your list of favorites
P10
09:00 - 18:00

EPOSTER - 10 Physics
Electronic Poster displayed from June 9 to June 13

09:00 - 18:00 #17763 - A validation method for SF QA through pinpoint ionization chamber measurement.
A validation method for SF QA through pinpoint ionization chamber measurement.

Introduction: Quality Assurance (QA) is a fundamental stage in radiotherapy treatment (RT), but When it comes to small-field (SF) irradiation, common in Stereotactic Radiation Surgeries (SRS), there is a lack of recommendations. Not until recently, with TRS483 (IEAE, September 2017), there was no document guiding SF-QA in a dosimetric perspective. The services that performed dosimetric QA based their methodology in articles published independently. In Brazil, the scarcity of guidelines is even more evident. Objective: We propose a method to implement dosimetric IF-QA of delivered dose calculated by the Treatment Planning System (TPS) and present cases of validation based on comparison to the CLINAC commissioning tables. Methodology: A cross-calibration was performed between two types of ionization chambers (IC), a referred-laboratory calibrated Farmer IC and a Pinpoint IC, in a standard reference field. The correction factor was obtained. The values for the correction by beam quality for small fields were then determined through the measure of TPR20,10 for different sizes of irradiation fields. After that, six SRS plans were reviewed, with treatment fields irradiating the Pinpoint CI. Readings were adjusted by the obtained KQ factors depending on TPR20,10, and the corrections of influence quantities were considered. Results: The values of TPR20,10 are coherent with those presented on TRS483, with a maximum difference of 2,5% for an irradiation field 4x4cm2. Among the reviewed treatment plans, the difference between the dose calculated by the TPS and the measured dose corrected by the factors varied from -2,7% to 1,3%. Discussion: The proposed methodology resulted in close values of estimated dose by the TPS and dose measured by the Pinpoint CI. The data of TPR20,10 and dose for small fields should be verified by Monte Carlo simulations for a more reliable validation. Conclusion: The proposed methodology is feasible, improving SRS treatment security through more reliable SF-QA.

André EZEQUIEL LÔBO DE ABREU, Tiago BATISTA OLIVERA (Ipatinga/MG, BRAZIL), Gerson Hiroshi YOSHINARI JR., Mariana PARANHOS ALVARENGA, Harley FRANCISCO DE OLIVEIRA
09:00 - 18:00 #17832 - Assessment of difference in dose volume histogram bin calculation for the new version of gamma knife treatment software.
Assessment of difference in dose volume histogram bin calculation for the new version of gamma knife treatment software.

Introduction: In the latest Gamma Knife planning software, Gamma Plan (GP) version 11.1.1, dose volume histograms (DVH) are now binned in 0.1 % at the measurements window compared to 0.5 % as it was before on the previous versions. The effect of this improved calculation can potentially change the prescription dose method especially for centers like ours where the 100% prescription coverage was usually defined using DVH from the measurement window. In this study we compared plans with 99.5 % coverage versus 99.95% coverage using the measurement window. The difference may not look large but it may potentially affect our goal for coverage in order to standardize our planning.

 

Methods: A total of 20 clinically used plans were created with 99.5% and 99.95% coverage by adjusting only the prescribed isodose lines (IDL) which was our standard clinical practice to achieve 100% coverage. Difference between both types of plans were evaluated using dosimetric treatment parameters as a voxel minimum dose, voxel maximum dose, maximum dose of 0.035 cm3 target volume dose, and a mean target dose.  The treatment plans differences in selectivity and gradient indexes were also studied.

 

Results: The increase for a voxel minimum dose, voxel maximum dose, maximum of 0.035 cm3 target volume dose and a mean target dose were in the range of 3 – 6% for 99.95 % coverage of the target compared with 99.5% of coverage.  As a result, by trying to cover an additional 0.45% of the target, we are effectively increasing the dose to the target for the same prescription. For instance, 15 Gy with 99.95% coverage could be more like 16 Gy with 99.5% coverage in terms of similar minimum, maximum and mean doses. Selectivity for the plans with 99.95% coverage decreased, but the gradient index improved.

 

Conclusion: The coverage goal will need to account for the difference in how DVHs are displayed in with the new GP update. This change may result in a clinically significant dose adjustment and needs to be considered for clinical implications and comparison with treatments delivered using old versions of GP.

Gennady NEYMAN (Cleveland, USA), Peng QI, Jennifer YU, Erin MURPHY, Gene BARNETT, John SUH, Samuel CHAO
09:00 - 18:00 #17563 - CBCT-guided cranial radiosurgery validation process by end-to-end test with TLD and film in a SRS head phantom.
CBCT-guided cranial radiosurgery validation process by end-to-end test with TLD and film in a SRS head phantom.

Purpose: The main purpose of this work was to show the availability of perform a frameless cone-beam computed tomography (CBCT)-guided cranial stereotatic radiosurgery (SRS) by an end-to-end (E2E) test with IROC Houston SRS head phantom using the Varian Clinac iX.

Methods and Materials: The head phantom has two inserts, one for the imaging containing the target and the dosimetry insert with film and thermoluminescent dosimeter (TLD). The imaging insert has a nylon sphere embedded in water as the target volume (simulating the tumor). The dosimetry insert contains two orthogonal sheets of Gafchromic passing throug the center of the "tumor" and two TLD within 0.5cm of the center of the target. The phantom with both inserts was imaged with computed tomography (CT) and magnetic resonance imaging (MRI), and a SRS dynamic arc plan was generated to the one with the imaging insert. The purpose of the plan was to cover the target with 25Gy with the prescription isodose being more than 85%. The phantom was localized at the treatment table with CBCT and the imaging insert was changed to the dosimetry one to be irradiated. Comparisons of the planned and the delivered dose to the film and TLD were performed. For the film, the coronal and sagital sheets were analysed with the gamma index with of difference dose 5% and distance-to-agreement (DTA) of 3mm with minimum of 85% of the points meeting the criteria. For the TLD the dose to the center of the target was compared with the plan and must meet the criteria of 95% to 105% of the read dose.

Results: The dosimetric results met the porposed criteria. The gamma index of the of both film sheets (coronal and sagital) agreed with the gamma index in 96% of points. The ratio of the dose to the center of the TLD and of the plan was 0.98.

Conclusions: The end-to-end test procedure proposed by to evaluate the abilities to locate and treat an intracranial target with high precision was performed with success in a frameless setup and CBCT method for positioning.

Anderson MELO (Maceió, BRAZIL), Landobergue BARROS
09:00 - 18:00 #17845 - Clinical implementation of a dedicated brain treatment planning optimizer for stereotactic treatments.
Clinical implementation of a dedicated brain treatment planning optimizer for stereotactic treatments.

Purpose:

Clinical implementation of a novel dedicated and automated treatment-planning solution for cranial indications, Elements. Single lesions can be targeted with an inversely optimized VMAT approach using automated arc trajectory optimization (Cranial SRS Element, Brainlab, München, Gernmany) while up to fifteen metastastic brain tumors can be automatically targeted with a single isocenter and multiple inversely-optimized dynamic conformal arcs (Multiple Brain Mets SRS Element, Brainlab, München, Gernmany).

Material and Methods:

The very first 25 treated patients were analyzed, each representing a variable number of lesions (1-12). Depending on the number and location of the lesions a dedicated Element was selected and used in order to achieve the specific planning constraints.The plans were evaluated by means of Paddick conformity (CI) and gradient index (GI). Patient specific quality assurance (QA) was performed with gafchromic EBT3 film and portal imager.

Results:

The Elements software tools generated plans with CI of 0.71±0.09 and a gradient index of 3.9±1.4. All plans achieved the organ at risk constraints. A gamma of 3%/3mm was used for the QA. A 98 % and 98,2 % passing rate was found for the EBT3 film and portal imager, respectively. This shows also the good concordance between film and EPID, suggesting that patient specific QA can be performed with the portal imager rather than the time-consuming films.

Conclusions:

The automated dose planning Elements revive dynamic conformal arcs as the paradigm for linac-based stereotactic radiosurgery of multiple brain metastases and at the same time implements an improved VMAT approach for single lesions with the use of automated arc trajectory optimization. This study shows the implementation of this technique in the routine clinical environment with an improved planning and treatment efficiency.

Thierry GEVAERT (Brussels, BELGIUM), Andrea GIRARDI, Benedikt ENGELS, Marlies BOUSSAER, Chaïmae EL AISATI, Mark DE RIDDER
09:00 - 18:00 #17897 - Commissioning and end-to-end validation for a high dose rate beam for radiosurgery.
Commissioning and end-to-end validation for a high dose rate beam for radiosurgery.

Currently, stereotactic radiosurgery and hypofractionated treatments become increasingly frequent, being the source of many studies. Associated with the new definitions and adjustments for small fields, the commissioning and validation of a high dose rate beam for small field treatments requires an adjustment to the factors suggested by the new IAEA document which infers factors for correction of the readings made by the detectors in the small field situations. A 6MV beam with high dose rate was commissioned and validated for Eclipse, with the AAA algorithm using the blue phantom, a PTW 3D pinpoint ionization chamber and an unshielded diode for the small fields. Following the protocol methodology, the 5x5 field was used as the fmsr field. Thus, the dose profiles: depth dose, inplane and crossplane of the beams were acquired with the two detectors for the square fields of 5 cm to 1 cm to identify the effects suggested by the literature. The output factor was acquired for all symmetrical and asymmetric field sizes with the pin point camera and, with the diode, for the fields equal to and less than 5 cm. Data were treated as recommended by the literature, including current work, and were fed into the planning system. The validation was done initially with a uniform phantom of solid water and already showed good agreement of the field factors and simplified plans, with discrepancies less than or equal to 1%. In an anthropomorphic phantom STEEV by cirs were simulated radiosurgery treatments with non-coplanar 3D and VMAT fields, both showed maximum discrepancies in approximately 1%.

Thiago B SILVEIRA, Maira R SANTOS (Rio de Janeiro, BRAZIL), Mozart G DUARTE
09:00 - 18:00 #17736 - Comparison between volumetric modulated arc therapy and dynamic conformal arc stereotactic radiotherapy for intracranial lesions.
Comparison between volumetric modulated arc therapy and dynamic conformal arc stereotactic radiotherapy for intracranial lesions.

The purpose of this study is to evaluate the dosimetric differences between treatment plans using dynamic conformal arc therapy (DCAT) from two TPS’s (Varian Eclipse and Brainlab elements) and volumetric modulated arc therapy (VMAT) from Varian Eclipse TPS, via frameless, linear accelerator based stereotactic radiotherapy (SRT) for the treatment of brain lesions for the first use of stereotactic radiotherapy in Algeria Using a Varian iX23 Linear Accelerator, Brainlab's 6D Exactrac Positioning System and Robotics couche, and TPS’s Elements (Brainlab) and Eclipse (Varian). Three plans were developed per each patient, with a total of fifteen patients, utilizing DCAT and VMAT. The plan comparisons of 45 treatment plans include the target coverage, conformity index (CI), homogeneity index (HI), gradient index (GI), and the volume of the normal brain tissue receiving doses of 12 and 5 Gray (Gy). Results of this research outline which planning method may provide benefits or lack thereof depending on the brain lesion location and size, thus providing data in terms of conformity of target coverage as well as lower dose spillage to the rest of the brain. This study also provides dosimetric results regarding advantages and disadvantages of forward versus inverse planning, in addition to the impact of a multi-leaf collimator (MLC) width size. Potentially the results of the study will indicate the most beneficial technique for delivery of SRS treatments for intracranial tumors.

Samir BENCHEIKH, Malika BESBAS, Ryma LOUELH, Mourad BELMESSAOUD, Abdelkader TOUTAOUI (Tizi Ouzou, ALGERIA)
09:00 - 18:00 #17726 - Contribution of scattered radiation to image formation and imaging dose in CyberKnife radiosurgery.
Contribution of scattered radiation to image formation and imaging dose in CyberKnife radiosurgery.

Purpose: Evaluate the contribution of scattered radiation to radiographic image formation and perform calculations of the imaging dose delivered by the x-ray based image guidance system of a CyberKnife radiosurgery system.

Methods: The CyberKnife image guidance subsystem consisting of two x-ray tubes and two flat panel detectors situated at the ceiling and the floor of the treatment room, respectively, were modeled and used to perform Monte Carlo (MC) simulations for water phantoms of different radii centered at the isocenter, as well as, for intracranial radiosurgery cases. Patient head models were simulated using lattice geometries constructed on the basis of information retrieved from corresponding CT scans. The MCNP6 general purpose MC code was employed to simulate photon transport and score i) the fluence of phantom/patient-scattered photons incident on the image detectors and ii) the imaging dose delivered for different kVp settings.

Results: The contribution of scattered radiation incident on the detector was found to depend on the dimensions of the imaging object and the kVp setting,reaching up to 30% at the center of the detectors for the 30cm diameter phantom and 150kVp. An imaging dose of 0.4mGy for the eye lenses and less than 0.1mGy for the healthy brain was calculated for an image acquisition using both tubes and typical settings of 120kVp and 10mAs.

Conclusions: The scatter contribution in radiographic image of a CyberKnife system is decreased due to the increased distance of 140cm between the patient and each detector. An imaging dose per acquisition of 0.4mGy was found for the eye lenses which corresponds to 4cGy for a treatment using a typical number of 100 acquisitions. Further studies to model the contribution of scattered radiation and improve low contrast resolution which is of interest in extracranial radiosurgery will be performed. 

Panagiotis ARCHONTAKIS, Argyris MOUTSATSOS, Emmanouil ZOROS (Greece, GREECE), Eleftherios PAPPAS, Evaggelos PANTELIS
09:00 - 18:00 #17788 - Development of a stereotactic system as a basic concept for diagnostic biopsy as well as lesioning or stimulation of relevant brain structures.
Development of a stereotactic system as a basic concept for diagnostic biopsy as well as lesioning or stimulation of relevant brain structures.

Currently, complex stereotactic systems are used to act in the brain as three-dimensional organ. The biggest barrier of these systems available to date is their high investment requirements (up to 150 TEURO), their enormous weight (up to 5000 g) and their very complex assembly, including the long learning curve.

The aim is to develop a safe and simple stereotactic system that guarantees the same or better precision as traditional systems, but greatly simplifies acquisition and handling.

We developed a system based on a patient-specific 3D platform, which contains a total weight of 300 g. This platform contains all relevant information of the target and entry point and allows a tenth of a millimeter placement of biopsy needle or functional equipment. First, three, max. four small bone screws (frontal/parietal) fixed in the cranial bone followed by the acquisition of a CT or MR dataset. With the planning data, we construct a virtual platform for placement of the desired instruments. This model is 3D printed and attached to the re-exposed bone screws during surgery. This offers the option of a bi-hemispheric operation and helps to reduce the cost per patient by up to 40% of the conventional price.

An initial accuracy study (n = 40) documents a precision of the instrument tip (biopsy needle) of 0.58 mm ± 0.34 mm (ranging from 0.09 mm to 1.17 mm), comparing the target performance based on CT data of planned to actual instruments tip.

The system developed in Germany represents an excellent alternative to traditional stereotactic systems and helps to establish stereotaxy as a routine procedure. Furthermore, these platforms are single use as well as individual for each patient. The implementation of all spatial coordinates in the design of the platform eliminates intraoperatively all adjustment processes and shortens the duration of surgery by up 30%.

Juan Carlos CAMACHO RODRÍGUEZ, Dirk WINKLER (, GERMANY), Christian FIEDLER, Robert MOEBIUS, Marcel MULLER, Ronny GRUNERT
09:00 - 18:00 #17795 - Evaluation of the Explorer 4D treatment planning system.
Evaluation of the Explorer 4D treatment planning system.

Objectives: The ARI GammaART 6000ND rotating gamma systemwas installed in August 2007 at the University of Debrecen and has been used to treat more than 4500 patients since then. Regular tests are performed by the medical physics group following stringent quality assurance guidelines.The objective of this paper is the assessment of the Explorer 4DTM  treatment planning system (TPS).

Methods:  GAFchromicTMEBT 3 films were used to verify the 50% isodose lines of the TPS. The films were irradiated in a water equivalent phantom and then evaluated with FilmQATMPro software. The actual results were compared with the 50% isodose lines of the TPS. A PTW pinpoint 3D ionization chamber was placed inside a water equivalent phantom to verify the output factors built into the TPS.

Results: Statistical analysis of the results show that the difference between the 50% isodose lines of the TPS and the actual irradiation was maximum 2.35%.  In regards to the output factors, the results show less than 1% deviation from the TPS, except for the 4mm collimator, where the measurements were not reliable.

Conclusions: The results show that the TPS performs within the required tolerances to perform stereotactic treatment planning for intracranial indications.

Tamás HOLLÓ, Gulyás LÁSZLÓ (Debrecen, HUNGARY), László BOGNÁR, Imre FEDORCSÁK, József Gábor DOBAI
09:00 - 18:00 #16810 - Four-dimensional digital tomosynthesis based on visual respiratory guidance.
Four-dimensional digital tomosynthesis based on visual respiratory guidance.

The aim of this research was to introduce and evaluate a respiratory-guided slow gantry rotation 4D digital tomosynthesis (DTS). For each of 10 volunteers, 2 breathing patterns were obtained for 3 minutes, one under free breathing condition and the other with visual respiratory-guidance using an in-house developed respiratory monitoring system based on pressure sensing. Visual guidance was performed using a 4s cycle sine wave with an amplitude corresponding to the average of end-inhalation peaks and end-exhalation valleys from the free-breathing pattern. The scan range was 40 degrees for each simulation, and the frame rate (FR) and gantry rotation speed (GRS) were determined so that one projection per phase should be included. Both acquisition time (AT) and the number of total projections to be acquired (NPA) were calculated. Applying the obtained respiration pattern and the corresponding sequence, virtual projections were acquired under a typical geometry of Varian on-board imager for two virtual phantoms, modified Shepp-Logan (mSL) and XCAT (extended Cardiac-Torso). For the XCAT, two different orientations were considered, anterior-posterior (i.e., coronal) and left-right (i.e., sagittal). Projections were sorted to 10 phases and image reconstruction was made using a modified filtered back-projection. Reconstructed images were compared with the planned breathing data (i.e., ideal situation) by SSIM (Structural Similarity) and NRMSE (Normalized Root-mean-square Error). For each case, simulation with guidance (SwG) showed motion-related artifact reduction compared to that under free-breathing (SuFB). SwG required less NPA but provided slightly higher SSIM and lower NRMSE values in all phantom images than SuFB did. In addition, the distribution of projections per phase was more regular in SwG. Through the proposed respiratory-guided 4D DTS, it is possible to reduce imaging dose while improving image quality.

Kim DONG-SU, Kim SIYONG, Suh TAE SUK (Seoul, KOREA)
09:00 - 18:00 #17719 - Frame-based to frameless brain SRS. Single centre experience in short time technological evolution.
Frame-based to frameless brain SRS. Single centre experience in short time technological evolution.

Mevaterapia medical centre put together a multidisciplinary team to implement frame-based brain SRS using a VARIAN Trilogy LINAC early 2017. Over 2 years a continuous technology upgrade has been performed in order to being able to perform safe frameless brain SRS.

First patient was treated during June 2017, with two treatment workflow considered; single fraction patients with stereotactic frame or hypofractionated patients with BrainLab stereotactic mask. Patient localization was performed based on CBCT imaging using a tolerance criteria 1mm/1°, achieving 6 degree positioning combining Varian couch with couch-mount movements.

During November 2017 ExacTrack imaging system was integrated to our LINAC, and imaging localization transition was carried out from CBCT to ExacTrac based imaging over 2 month. Positioning tolerance criteria was kept 1mm/1°.

In May 2018, we upgrade our treatment couch to BrainLab robotic 6D couch allowing ourselves to perform frameless SRS treatments. Frame-based to Frameless workflow was carried out progressively over 4 month with selected patients in joint decision between radiation oncologist and neurosurgeons. Multiple brain metastasis SRS was incorporated to our clinical protocol and positioning tolerance was set to 0.5mm/0.5°.

In room treatment time over our technological evolution was assessed and compared, considering that positioning tolerance was kept over these 2 years and more than 100 patients. CBCT imaging based treatments were performed in 60±20 minutes, when ExacTrac was implemented with Varian couch total time reduced to 40±10 minutes. Finally, actual in room treatment time with Frameless technology using ExacTrac plus robotic 6D couch is reduced to 30±10 minutes. Total patient time in treatment was reduced in over 40% with equivalent treatment quality and a quantitative increase in patient comfortability.

Florencia MAURI, Leon ALDROVANDI, Ruben Oscar FARIAS, Augusto ALVA, Pablo Marcelo AJLER, Matteo BACCANELLI, Federico Javier DIAZ, Maria Liliana MAIRAL, Claudio Gustavo YAMPOLSKY, Mabel Edith SARDI, Mara Lia SCARABINO (Buenos Aires, ARGENTINA)
09:00 - 18:00 #17566 - Geant4-based Monte Carlo simulation of correction factors for reference dosimetry of the Leksell Gamma Knife Perfexion.
Geant4-based Monte Carlo simulation of correction factors for reference dosimetry of the Leksell Gamma Knife Perfexion.

With the publication of TRS-483 in late 2017 the IAEA has established an international Code of Practice (COP) for reference dosimetry in small and non-standard fields based on a formalism first suggested by Alfonso et al. in 2008. However, data on correction factors for the Leksell Gamma Knife Perfexion is scarce and what little data is available was obtained under conditions not necessarily in accordance with the IAEA’s recommendations. This study constitutes the first systematic attempt to calculate those correction factors by applying the new COP to Monte Carlo simulation using the GEANT4 toolkit. The correction factors were determined for three common ionization chamber detectors, modeled in great detail based on proprietary blueprints provided by their respective manufacturers, in five different phantom materials and using three different physics lists. The results indicated that for chambers with a collector electrode made of low-Z materials, correction factors were within 1% of unity for the liquid water, Solid Water™ and polystyrene phantom materials, whereas chambers with a collector electrode made of heavier elements and electronically denser phantom materials necessitated larger corrections. The correction factors did not differ significantly between the various physics lists. Similarities and differences between the results of this study and previous ones based on EGSnrc and PENELOPE-based Monte Carlo codes were also analyzed and it was found that the results obtained herein were generally in good agreement with the findings of earlier studies that were obtained under comparable reference conditions. The correction factors obtained in this study can be a contribution to a potential revision or update of TRS-483.

Thomas SCHAARSCHMIDT, Tae-Hoon KIM, Young Kyun KIM, Hye Jeong YANG, Kook Jin CHUN, Eun Young KIM, Hyun-Tai CHUNG (Seoul, KOREA)
09:00 - 18:00 #17890 - Hypofractionated treatments using Gamma Knife Icon: accuracy evaluation of the daily repositioning.
Hypofractionated treatments using Gamma Knife Icon: accuracy evaluation of the daily repositioning.

The major improvement of Gamma Knife Icon is the possibility to execute frameless treatments to treat large-volume lesions or lesions close to critical organs reducing the side effects to normal tissues.

A possible issue using the frameless modality for hypofractionated treatments is the accuracy and reproducibility of the patient positioning. 

The patient displacement measurements (rotation and translation along the x/y/z axis) have been collected for each fraction and then the data have been analyzed to verify the accuracy and the efficiency of the current positioning procedure.

Between October 2017 and January 2019, 417 patients were treated in Fondazione Poliambulanza, including 65 in frameless mode with the thermoplastic mask, in single or multi-session fractions. In the first 57 hypofractionated treatments (3,4,5 days, for a total of 256 fractions) we studied the error (offset) in the daily repositioning of the patient in the three X-axis, Y, Z both for rotation and for translation. We obtained the following results (the value can be positive or negative depending on the side from which the new positioning moves with respect to the first day): 

- X rotation: mean -0.10°, SD 0.62°; 

- Y rotation: mean 0.08°, SD.53°; 

- Z rotation: mean 0.05°, SD 0.40°; 

- Translation X: mean -0.02 mm, SD 0.25 mm; 

- Translation Y: mean -0.28 mm, SD 0.22 mm; 

- Translation Z: mean -0.04 mm, SD 0.99 mm. 

The most frequent error in translations was found within 0.5 mm; in rotations, within 1°. In translations, the axis most frequently involved is the Y axis (antero-posterior), while the higher error, although less frequent, occurs above all on the Z axis (head-feet). For the rotations, the statistic did not show a preferential axis between X, Y or Z. Overall, the fixing system consisting of the thermoplastic mask and the customized conformable pillow was reliable, easy to use for the operator and was well tolerated by all the treated patients.

Alberto FRANZIN (Brescia, ITALY), Chiara BASSETTI, Lodoviga GIUDICE, Cesare GIORGI, Ivan VILLA, Marco GALELLI, Oscar VIVALDI, Mario BIGNARDI
09:00 - 18:00 #17880 - Image processing for radiosurgery using computer deep learning.
Image processing for radiosurgery using computer deep learning.

In radiosurgical operations, the accuracy of planning and drawing of target structures as well as the definition of risk structures is very important.

The planning of the radiosurgery is performed mainly by CT and MRI images, which are transferred after they are merged into the virtual planning system. Visualization and virtual planning greatly assist the subjectivity of hand drawing contour of brain structures by the neurosurgeon for optimization of treatment.

The new proposed system allows to search quickly for the very large amount of information needed to better diagnose and design a further treatment that help the physician in the decision-making algorithm and provide better treatment outcomes for radiosurgery. Priority is given to the processing of image information from CT, MRI and PET considering the different quality of the devices, which is reflected in the different sensitivity and specificity of the results obtained by imaging and clinical methods. This information, if necessary, are displayed in 3D stereoscopy. The processing of sophisticated information system methods by combining the results of multiple imaging (CT, MRI, PET) increase the specificity and sensitivity of the algorithm of interpretation and decision making for the diagnosis and treatment of the patient, which is also a shift towards personalization for the decision algorithm. This system allows through the secure network to communicate between hospitals to diagnose correctly and thus decide the treatment method correctly. At the same time, the system is designed to have the ability to communicate with existing PACS systems in hospitals and "self-deep learning" skills.

Visualization and communication across the hospital network will enable you to prepare radiosurgery with experts from different hospitals, thus increasing the emphasis on the accuracy of the planning process.

Miron ŠRAMKA, Eugen RUŽICKÝ (Bratislava, SLOVAKIA), Alena FURDOVÁ, Štefan KOZÁK, Ján LACKO
09:00 - 18:00 #17842 - Impact of patient-specific MRI distortion correction for stereotactic cranial target definition.
Impact of patient-specific MRI distortion correction for stereotactic cranial target definition.

Introduction

The accuracy of a stereotactic treatment is primarily limited by the least accurate process in the whole chain of events from patient scanning to patient treatment.

The targeting is limited by the accuracy of the CT and MRI images. MRI datasets are subjected to distortions, due to nonlinearity of gradient fields, andmay cause incorrect target definition.

This study aimed to analyze the impact of a patient-specific algorithm, Crainial distortion Elements (Brainlab, München, Germany), rather than a manufacture-specific, to correct spatial distortion in cranial magnetic resonance images.

Methods and materials

Twelve trigeminal patients treated with a single dose of 90 Gy with a 4mm collimator were studied retrospectively.  The gross target volume (GTV) was defined on a 1.0mm T1 MPRAGE and T2 MRI corrected for distortion with a machine-specific algorithm.

For this study, the manufacture-specific corrected MRI was further corrected using a patient-specific distortion correction algorithm that references the treatment planning CT. The GTV were then mapped onto this newly created patient specific corrected MRI dataset.

The original defined target and the corrected deformed object were mutually compared by means of several quantitative measures such as Dice, Jaccard, and Hausdorff indices. The average distance between the two centers of the two GTV was also calculated.

 

Results

On average, a good agreement was found between both GTV resulting in a Dice index of 0.76 (SD 0.23) ranging between 0.13 and 0.92. The Jaccard index, which is an intersection over Union was similar (p> 0.1) to the Dice with an average of 0.66 (SD 0.23) ranging between 0.09 and 0.86. The greatest of all the distances from a point in GTV to the closest point in the other GTV, called the Hausdorff distance, was 0.73 on average (range 0.50-1.80), reflecting good similarity between both GTVs.Average distance between both GTV was 0.43 mm (SD0.26mm), with a minimum of 0.20 mm and a maximum of 1.10 mm. One out of the 12 patients met criteria of “geometric miss”, which was not correlated with clinical outcome.

 

Conclusion

Our study showed that the cranial distortion Elements correct all images even when manufacture-specific corrections fail due to patient specific conditions. In order to avoid any geometrical miss, a patient specific distortion correction must be applied for all cranial indication.

Thierry GEVAERT (Brussels, BELGIUM), Benedikt ENGELS, Chaïmae EL AISATI, Mark DE RIDDER
09:00 - 18:00 #17751 - Implementation of a national stereotactic radiosurgery chart round: engagement and clinically significant outcomes.
Implementation of a national stereotactic radiosurgery chart round: engagement and clinically significant outcomes.

Introduction

Intracranial and extracranial Stereotactic Radiosurgery (SRS) are highly specialised techniques which require careful and structured implementation.  Here we describe the implementation of a weekly national SRS chart round attended by specialists from 10 geographical locations, held via videoconference. Prior to SRS delivery, all patients had their case presented where clinical, planning and technical details were discussed.

 

Method

Data was collected from weekly run chart rounds between July 2018 and January 2019.  Details recorded included specialty attendances, clinical background, diagnosis, volumes, radiation dose and fractionation, treatment site/location and clinician approach.  Furthermore, consensus recommendations regarding changes to treatment approaches were also recorded.

 

Results

There were 590 attendances split across radiation therapists/dosimetrists (29.0%), radiation medical physicists (27.6%), oncologists (27.3%) and management (16.1%).  There were 118 cases presented. Of these, primary or oligometastatic lung malignancies represented the majority of cases (n=42), followed by brain metastases (n=21 and n=14 for single met and multi-met respectively), bony oligometastases (n=17), non-bone oligometastases (n=11), spine (n=3), benign brain conditions (n=3) and primary brain (n=1).  Across our national network we have had 18 radiation oncologists present at least one case at the chart round with others participating for training and mentorship.  Of the 118 cases presented, there were 29 (24.6%) recommendations made for 21 patients.  These included changes to contours (n=10), dosimetry (n=4), treatment technique (n=1) dose/prescription (n=8) and 6 occasions where standard fractionation or no treatment was recommended.  Three patients had more than one recommendation made.  Cases were re-presented following changes from recommendations.  

 

Conclusion

The implementation of a national SRS chart round, held via videoconference has ensured national protocol compliance to stereotactic treatments across our network.  Furthermore, the chart rounds have allowed clinicians to be provided with mentorship and guidance from nationally and internationally recognised SRS experts which has allowed for increased plan quality and patient outcomes.

Rhys FITZGERALD (Brisbane, AUSTRALIA), Trent ALAND, David PRYOR, Lee ANDERSON, Andrew FONG, Dominic LUNN, Andrew OAR, Marcel KNESL, Jim JACKSON, Matthew FOOTE
09:00 - 18:00 #17710 - Implementation of a novel non-coplanar arcs technique for stereotactic treatment of brain metastases.
Implementation of a novel non-coplanar arcs technique for stereotactic treatment of brain metastases.

Purpose: Implementation of stereotactic radiosurgery (SRS) treatment for brain metastases in our department with no previous SRS experience, using the dedicated technique HyperArc™, Varian Inc. (HA) on a TrueBeam STx.

Material and Methods: The accuracy of the HA patient positioning system, MV and Cone-Beam computed tomography (CBCT) in conjunction with the 6D robotic couch were evaluated. Brainlab ExacTrac (ET) was used to verify patient positioning prior to each beam and trigger new CBCT correction. Dedicated beam models for millennium and HD multi leaf collimator (MLC) were commissioned. Coincidence of imaging and radiation isocenters was confirmed using Winston-Lutz (WL) test. EBT3 film measurements were done for Picket Fence (PF) test, end-to-end (EE) test using a Max HD phantom and clinical pre-treatment Quality-Control (QC) using a CIRS phantom. The first 21 patients (1-4 brain metastases) have been treated using 4 non-coplanar 180°-arcs in one fraction of 18Gy. A 2 mm margin was applied to the gross target volume delineated on magnetic resonance images. Plans were evaluated with dosimetric indices, total monitor units and overall treatment time (OTT) per fraction.

Results: The CBCT and ET isocenters agreed to within 0.4 mm and 0.3° in translational and rotational directions. The beam models were accurate down to a jaw field of 2 x 2 cm2 with dosimetric leaf gaps of 0.15 cm and 0.09 cm and transmission factors of 1.8% and 1.2% for 6 MV flattening filter free photon beam with the millennium and HD MLC, respectively. The results of WL (max delta <0.7mm), PF and EE (gamma passing rate >95%) tests were within the defined criteria. The mean Planning Target Volume was 3.3 cm3 (range 0.3-16 cm3). Dose constraints were within tolerances for all the patients. Pre-treatment QC resulted in a local gamma passing rate (1mm/5%) above 90% for all the patients. Maximum observed deviation in patient positioning was 3.2mm/2.4°, 11 patients were within treatment tolerance (1mm/1°) throughout treatment. The mean OTT was 33.5 ± 13.5 min.

Conclusion: HA offers fast accurate treatment planning and dose delivery. By omitting MV images, OTT can be decreased by almost half. Next steps are to investigate the possibility of using only the initial CBCT as image verification, and possible benefits of using HD MLC.

Lucie CALMELS (Herlev, DENMARK), Susan BLAK NYRUP BIANCARDO, David MC KENZIE GRANT, Maria SJÖLIN, Eva WILKEN, Sune Kristian BUHL, Ulf BJELKENGREN, Mette ANDERSEN, Patrik SIBOLT, Mette PEDERSEN, Hanne LANDGREN, Susanne LIND, Anja NIELSEN, Poul GEERTSEN, David SJÖSTRÖM
09:00 - 18:00 #17722 - Initial experience using a bolus skin-equivalent for calvarial metastases and skull defects using the Leksell Gamma Knife ICON mask based system: The Cleveland Clinic Experience.
Initial experience using a bolus skin-equivalent for calvarial metastases and skull defects using the Leksell Gamma Knife ICON mask based system: The Cleveland Clinic Experience.

Introduction: Our group has previously published on the use of a bolus skin-equivalent layer for treating calvarial and skull base metastases using the Perfexion frame based system (Kotecha et al J NSG (Suppl 2) 121:91-101, 2014). Traditionally used TMR 10 algorithms in Gamma Knife Radiosurgery (GKRS) are not accurate in the first 5 mm from the surface and by adding at least 5 mm of bolus material to the mask and creating an extended skull contour that limitation is removed.

Method: We used a bolus skin-equivalent attached directly to the Leksell Gamma Knife ICON thermoplastic mesh faced mask. The system uses image-guidance utilizing cone-beam CT (CBCT) and infrared tracking to ensure minimal inter- and intra-fractional movement during GKRS and hence can also be used in a fractionated manner. Four patients to date have been treated. The thermoplastic mask system consisted of a customized resin-filled neck rest and a 3-point thermoplastic mask. The area to be treated was outlined on the skin so the correct size of bolus would be utilized. A bolus skin-equivalent of at least 5 mm was positioned over the treatment site and attached directly to the thermoplastic mask to artificially extend the surface to target distance. A localization CBCT was completed after the mask was made and the patients then completed high resolution (1 mm slice) contrasted enhanced MRI images and CT scans without and coregistered to reduce inaccuracies from image distortion.

Results: The four cases treated included a 14 year old (y/o) with osteosarcoma skull metastases treated over 3 fractions, a 69 y/o with renal cancer skull metastases treated over 5 fractions, an 81 y/o with atypical meningioma and skull defect treated over 5 fractions, and a 55 y/o with an anaplastic oligodendroglioma and a skull defect treated in a single fraction. The four cases will be presented in detail outlining the treatment flow. We were able to calculate the skin max point dose (Gy) for 3 patients with all falling below the skin tissue constraints using the Timmerman Tables. There was associated hair loss but no skin dermatitis.

Conclusion: Gamma Knife using the ICON mask based system and a bolus allows treatment of superficial calvarial lesions and patients with superficial tumors and skull defects and provides an option for patients who are not candidates for a frame based Gamma Knife procedure.

Glen STEVENS (Cleveland, USA), Lilyana ANGELOV, Sam CHAO, Gennady NEYMAN, Erin MURPHY, Dani FLAK, John SUH
09:00 - 18:00 #17672 - Integrating navigated Transcranial Magnetic Stimulation (n-TMS) in gamma knife radiosurgery planning.
Integrating navigated Transcranial Magnetic Stimulation (n-TMS) in gamma knife radiosurgery planning.

Background:

To illustrate how navigated transcranial magnetic simulation (n-TMS) can be utilized in the radiosurgical management of brain metastases involving areas of the motor cortex.

Case Descriptions:

Case 1: A 53 year-old woman with metastatic breast cancer developed focal epileptic seizures and weakness in her left hand.  A magnetic resonance imaging (MRI) scan demonstrated a partially cystic 30 mm metastasis in the right precentral gyrus and central sulcus. The lesion was treated with adaptive hypofractionated gamma knife radiosurgery; nTMS - based motor mapping was performed prior to treatment. Follow-up MRI up to 12 months revealed a significant decrease in tumor size without adverse radiation effects (ARE); symptoms resolved within one month post treatment.

Case 2: A 73-year-old man with metastatic lung cancer developed left hand weakness. The corresponding MRI demonstrated a 26 mm metastasis in the right postcentral gyrus and sulcus, 5 mm from the hand motor cortex. The patient underwent preoperative nTMS  motor mapping  prior to single dose gamma knife radiosurgery for both lesions. Follow-up MRI examinations up to 10 months showed tumor control and evolving ARE. Despite the latter, the patient experienced motor function improvement  during follow-up.

Conclusion: 

In our case series, nTMS was safely and effectively integrated in gamma knife radiosurgery (GKRS) planning. Motor mapping allowed sparing of healthy functional tissue. The relation between the type of radiation schedule and possible radiation-induced focal plastic distortions at long term needs consideration and deeper analysis.  Prospective studies involving larger, homogenous group of patients are warranted to further validate the clinical significance of nTMS in GKRS-treatment planning. 

Georges SINCLAIR, Georges SINCLAIR (Reading, UK, UK), Georges SINCLAIR, Gerald COORAY, Hamza BENMAKHLOUF , Christer LINDQUIST, Mominul ISLAM
09:00 - 18:00 #17811 - Monte Carlo evaluation of the effect of source self-shielding on Gamma Knife dose rate variation.
Monte Carlo evaluation of the effect of source self-shielding on Gamma Knife dose rate variation.

In this work, we analyze the effects of source self-attenuation on the dose rate for a Model C Gamma Knife unit (Elekta, Stockholm. As each source is composed of 20, 1 mm thick individual pellets and the source housing necessarily has a diameter greater than the pellets, it is possible that these pellets can move independently relative to each other. By modeling a large number of different random arrangements of pellets using the Monte Carlo code MCNP6, we were able to construct a probability distribution of the possible dose rates. Using a maximum pellet displacement off of the centerline of 0.25 mm (which assumes the source housing diameter is 0.5 mm greater than the pellet diameter), the FWHM of this distribution is 2.3%. Perhaps more interestingly, the mean dose rate of this distribution is 8% higher than the dose rate obtained when all of the pellets are perfectly aligned, indicating that while randomized arrangements do not vary by much in dose rate, allowing for some offset (and subsequently decreasing the source self-attenuation) can cause an appreciable increase in dose rate.

We then looked at using 2 mm diameter source pellets to match the Gamma Knife hole size, which reduced the number of pellets needed from 20 to 5 (changing the source height from 20 to 5 mm). This change led to a 32.5% increase in dose rate at isocenter for the 18 mm collimator. As we do not have an accurate model of the PerflexionTM system at the time of writing, we were not able to similarly evaluate the newest Gamma Knife generation. However, based on these results, we believe it is worthwhile to investigate the feasibility of modifying the Gamma Knife sources by reducing their dimensions along the beam line in order to reduce the effects of self-attenuation.

Gregory SZALKOWSKI, Tanxia QU (New York, USA), C-K Chris WANG
09:00 - 18:00 #17723 - Patient specific dose verification using a phantom duplicating the patient anatomy created with 3D-printing technology.
Patient specific dose verification using a phantom duplicating the patient anatomy created with 3D-printing technology.

Objective: The aim of this work is to establish and implement a patient specific end to end quality assurance methodology for dose verification in advanced radiotherapy applications using 3D-printing technology.

Methods: Eleven patient VMAT plans including either stereotactic or re-irradiation cases of primary or recurrent brain or head and neck tumors, created in Monaco TPS, were verified. A 3D-printer was used to construct a hollow phantom that duplicates the patient anatomical geometry, including bone structures, using the patients' planning-CT DICOM images. Special inserts were also constructed to position either a semiflex PTW (volume:0.125 cc) or a CC01 IBA (volume:0.01 cc) ionization chamber PTV and OARs. The hollow phantom was subsequently filled with water to simulate normal brain. The phantom was irradiated using the specific patient’s irradiation protocol including the IGRT step using CBCT with Elekta HexaPOD 6D robotic couch.  Phantom CBCT- images were co-registered with the patient-CT images in the TPS to accurately define ionization chamber positions. TPS calculations in the patient anatomy were calculated and compared with corresponding measurements.

Results: An excellent agreement (difference <4%, average=2.7%±1.2%) between measurements and TPS calculations were observed in the low dose gradient/high dose region within PTV. In the OARs region, the degree of agreement between measurements and calculations depend significantly on the definition accuracy of the position of the ionization chamber volume in patient anatomy as well as on the dose gradient in the region of measurement with differences being increased in the high dose gradient regions. In any case, differences < 10% (average=5.4%±2.4%) were observed with the measured dose being always lower than the accepted dose limit for the specific OAR.

Conclusion: The implemented methodology based on 3D-printing technology was found capable to verify the dose in clinically significant regions within the patient without the need of plan recalculation in the phantom anatomy.

 

Nikolaos GIAKOUMAKIS, Pantelis KARAISKOS (Athens, GREECE), Chryssa PARASKEVOPOULOU , Efi KOUTSOUVELI, Georgios KRITSELIS, Georgios KOLLIAS
09:00 - 18:00 #17846 - Polyetheretherketone (PEEK) implant can reduce postoperative artefacts and improve accuracy rate of delineation : a test in both pig and human.
Polyetheretherketone (PEEK) implant can reduce postoperative artefacts and improve accuracy rate of delineation : a test in both pig and human.

Background

Spinal stereotactic body radiotherapy (SBRT) delivers high doses of radiation and it is highly conformally focusing the radiation dose on the metastatic bone while sparing spinal cord. Postoperative SBRT dose planning relies on CT and MRI imaging. Polyetheretherketone (PEEK) is a new material of radiolucent character.

Aim is to compare the artefacts in CT and MRI scans caused by PEEK and Titanium rods implanted in pigs and humans.

Methods

In the pig spine specimen three groups of implants were sequentially inserted: a) Two Titanium rods, b) one Titanium rod and one PEEK rod, c) two PEEK rods. CT and MRI scans were acquired of all groups. A region of interest (ROI) was defined in order to measure the imaging noise caused by the rods. CT Houndsfield units (HU) were measured in ROIs and the image-noise were compared by calculating  artefact density standard deviation (SD).

The accuracy of spinal cord identification on MRI scans was compared on two patients who underwent spinal stabilisation.

Results

In the CT scans of pig specimen, the image-noise (artefact density standard deviation) was 63.5 HU for the titanium rods and 6.2 HU for the PEEK rods. There was a significant difference in image-noise between the two groups (P<0.01). The artefacts in the CT scans caused by the implanted rods were considerably lower for PEEK than for Titanium. For the two patients with respectively implanted PEEK and Titanium rods, it was only possible to identify the spinal cord for the patient with PEEK implants.

 Conclusions

PEEK rods created significantly less artefacts than Titanium rods in both CT and MRI scans, thus enabling more accurate spinal cord definition before SBRT. By using this new material, patients could benefit from a more precise and secure SBRT treatment  with less risk of radiation induced side-effects.

Miao WANG (Aarhus, DENMARK), Yasmin LASSEN-RAMSHAD, Esben Schjødt WORM, Simon Toftgaard SKOV, Haisheng LI, Anja HARBØLL, Kristian HØY, Ming SUN, Akmal SAFWAT, Ebbe Stender HANSEN, Kestutis VALANCIUS, Simon BUSS, Lise Nørgaard BENTZEN, Thomas BENDER, Morten HØYER, Cody BÜNGER
09:00 - 18:00 #17826 - Spontaneous intraocular air arising after placement of stereotactic frame.
Spontaneous intraocular air arising after placement of stereotactic frame.

Introduction: Although used since the last century, its application still can bring unexpected findings. It can be challenging after a craniotomy, requiring frame rotation and/or pins adjustment to avoid the bone flap. 

Objective: To report the spontaneous appearance of intraocular air immediately after stereotactic frame placement. 

Methods: A 61-year-old woman diagnosed with a left frontal tumor underwent microsurgery, confirming a fibrous meningioma. After complete resection, the tumor relapsed 4 years later. She elected to be treated with radiosurgery. A Leksell model G frame (Elekta, Sweden) was applied with fixation in a more basal manner than usual to avoid the bone flap of the previous left frontal craniotomy. She had eyelid edema immediately after injection of local anesthetic on the right eye. Immediate CT scan after the frame fixation showed the presence of intraocular air on the left eye (same side of craniotomy). This finding was absent on the MRI taken for planning a day earlier. Rigorous assessment did not show a bone fracture or pin slid into the orbit that would justify this finding. Ophthalmological infection was also ruled out. Radiosurgery underwent uneventfully, using a prescription of 15 Gy to the 50% isodose line. The patient reported a scotoma on the following day after radiosurgery. CT scan was repeated 8 days after GK showing complete resolution of intraocular air. Nevertheless, the patient continued to report a blind spot, later confirmed with a visual field exam. The patient was evaluated by a neuro-ophthalmologist who diagnosed a significant retinal pathology on the left eye. Intraocular air occurred ipsilaterally to the tumor and the craniotomy.

Conclusion: This is an unlikely concomitant occurrence of intraocular air accompanied by scotoma following stereotactic frame placement. Neuro-ophthalmologic investigation confirmed the co-existence of underdiagnosed retinopathy.

Juliete MELO DINIZ (SAO PAULO, BRAZIL), Antônio DE SALLES, Rafael COSTA LIMA MAIA, Aline Lariessy CAMPOS PAIVA, Bruno Henrique DALLO GALLO, Crystian WILIAN CHAGAS SARAIVA, Alessandra GORGULHO
09:00 - 18:00 #17717 - Stereotactic frame-based registrations methods for the Leksell Gamma Knife®.
Stereotactic frame-based registrations methods for the Leksell Gamma Knife®.

Introduction: The Leksell G-frame is designed to localize and immobilize the patient during imaging and treatment with e.g. the Leksell Gamma Knife®. The localization system uses the indicator box with N-shaped fiducials, to create a stereotactic space for tomographic studies used for registration in the Leksell GammaPlan®(LGP). Two new registration algorithms are proposed and evaluated on several tomographic image studies. The purpose is to investigate the registration differences between the LGP registration algorithm and the proposed registration algorithms in images where varying amount of distortion is present.

Materials and methods: 25 CT and MRI studies were stereotactically defined in a Leksell Gamma Plan 11.1 research version from which fiducial information could be exported for evaluation. LGP registers the corners of the stereotactic N derived from least-square fit of fiducials whereas the two proposed algorithms use the fiducial coordinates to minimize the distance of the fiducial to the theoretical fiducial N, either with respect to the mean or quadratic fiducial registration distance, which handles distortions differently. Differences in fiducial registrations and coordinates in stereotactic space were evaluated.

Results: The proposed registration methods proved stable and resulted in smaller mean fiducial distances compared to LGP; Mean Value method MR: 0.479±0.082mm, CT: 0.176±0.047mm; Quadratic method MR 0.484±0.081mm, CT: 0.179±0.050mm compared to LGP MR 0.559±0.088mm, CT 0.353±0.145mm. The quadratic formulation suppresses noise efficiently while the mean formulation reduces the effects of systematic fiducial deviations. The mean and max difference in stereotactical coordinates between registrations based on LGP vs the proposed formulations were 0.4mm and 0.9mm, respectively.

Conclusion: The small shift may be insignificant for most of the image sets analyzed. Larger deviations with LGP registrations (2D error) gave transformations resulting in larger deviation in the stereotactic volume. Further investigations are needed to evaluate stereotactic definitions.

Jonas JOHANSSON (Stockholm, SWEDEN)
09:00 - 18:00 #17865 - Tears: A Bizarre Cause of Collision in Gamma Knife Radiosurgery.
Tears: A Bizarre Cause of Collision in Gamma Knife Radiosurgery.

Background: With the availability of Perfexion and the Icon model of Gamma knife radiosurgery, no part of the brain is an exception for the treatment. 

Objective: For all practical purposes, the most common cause of the collision is not the patient’s head but the post or screw. Collision warnings distribute mostly at the anterior and the posterior ends of the head and less in the cranio-caudal direction. There is no other plausible source of collision in the radiosurgery. 

Case presentation: We have faced a difficult situation while administering gamma knife treatment to a 7-year-old boy with residual resistant acromegaly. The frame was placed equidistant keeping the nose in the center thereby centralizing the sellar region. All collision checks were confirmed, and no collision was reported, as the lesion was central and easily achieved with the treatment plan. Shortly after starting the treatment, the child got restless and started crying. After a while, the machine stopped the treatment indicating a collision error, but we could not find any source of contact.

Conclusion: The cone inside the treatment area is a collision sensor suspended on sensitive springs. The sensors’ function is to suspend treatment if the cone is displaced such that it creates an electrical connection with the collimator system. This shorting to the ground the trigger to prevent a mechanical collision between the fixated patient and the collimator. The sources are withdrawn to the park position and treatment stops. It seems that the patient’s tears, in this case, were sufficient to trigger the collision sensor. One can wonder if the system has become overly sensitive, in which case it would be recommended to inspect and perhaps change the springs. This case highlights the extreme sensitivity of the gamma gantry to avoid any unnecessary radiation or wrong delivery of the treatment. The treatment of pediatric patients under general anesthesia is another solution for this possible complication.

Manjul TRIPATHI (Chandigarh, INDIA)
09:00 - 18:00 #17654 - The reason for complying with the treatment time limit in plan competition for GKRS.
The reason for complying with the treatment time limit in plan competition for GKRS.

With the purpose of improving the quality of radiosurgery, there were for the first time two treatment plan competitions of brain stereotactic radiosurgery in 2018. They both have 5 target, prescription dose of 20Gy, and treatment time limitation in common. At the last minute 1st plan competition ignored the treatment time limitation on request from the users but 2nd plan competition kept it until the end. We analyzed the result of two competitions and figured out that the decision of keeping the treatment time limit or not was a key to ranking in GKS. ProKnow, 1st competition’s organizer, announced the result on website. It selected the top 50 among all participants with the individual information, acquisition score and the summary of the treatment plan. I analyzed the 7 people’s treatment plan registered in top 50 from the website and added one more plan that was one of my 2 treatment plans presented to the competition but not ranked in it. The result of RadiationKnowledge, 2nd competition’s organizer, took a little more time to be announced on the website and had only one gamma knife treatment plan in top 20 giving only individual information and acquisition score. I could review the plan information because the plan was mine. The first competition did not give the information of the treatment time but give the number of shots instead. We could anticipate the treatment time from the 3 plans that the owner of the plan had sent to me. Among 8 plans only one plan with 18 shots and 103 minutes kept the treatment time limitation. Seeing that the treatment plan with 26 shots had the treatment time of 146 minutes, the others must have exceeded the time limit of 120 minutes. The second competition must have kept the time limit of 90 minutes. The plan only recorded in the top 20 had the treatment time of 89 minutes with 19 shots. Even though the plans taking longer treatment time were certainly GKS planning for clinical treatment, the time limit should have kept to fairly compete in the game and make a right order in ranking. It seemed that the plan competition was to evaluate the ability to meet the condition not to find the highest score. If the first competition’s organizer had commented about this, the participants would have noted that they should comply with the time limit from next plan competition, such as RadiationKnowledge's plan competition.

Weon-Seop SEO (SEOUL, KOREA), Chang-Kyu PARK, Seok-Keun CHOI, Bong-Jin PARK
09:00 - 18:00 #17745 - The use of correction factor free detectors to validate ultra-small field dose distributions.
The use of correction factor free detectors to validate ultra-small field dose distributions.

Purpose: Ultra-small field dosimetry (<1cm) is challenging due to perturbation effects and 
volume averaging. Recently, small field correction factors have been introduced to compensate 
for the the varying responses of different detectors in radiosurgical fields. We demonstrate that 
data acquired and validated with detectors not requiring correction factors, such as the W1 
(Standard Imaging, Madison, WI) and Gafchromic film (Ashland, Bridgewater, NJ), yields 
excellent agreement between plans created in the Multiple Met (MME) Element (Brainlab, 
Munich, Germany) and phantom measurements. 
 
Methods: Output factors were acquired on a Truebeam STX (Varian Medical Systems, Palo 
Alto, CA) with a flattening filter free 6 MV beam using an Exradin W1 scintillator. Treatment 
plans for 41 targets were validated in MME: 5 with the W1 and 36 with Gafchromic EBT3 or XD 
film. Target sizes ranged from 0.06 cc to 0.6 cc and 0.03 cc to 1.7 cc for the W1 and film 
validations, respectively. Treatment plans were planned and delivered on MaxHD (IMT, Troy, 
NY) and Baby Blue (Standard Imaging, Madison, WI) phantoms. Phantom alignment was 
performed before and during delivery using Brainlab’s ExacTrac with thresholds of 0.5mm and 
0.5°. 
 
Results: All plans validated with film and the W1 had good agreement with calculations. 
Gamma scores >96% (2%/1mm, 10% dose threshold) were achieved for all film validations for 
targets ranging from 0.03cc to 1.7cc. W1 point dose measurements were <1.5% of calculation 
for all but the 0.06cc target, at which the target size is comparable to the W1 dimensions.  
 
Conclusions: Using a detector with a unity correction factor (or applying proper correction 
factors to an appropriate detector, such as those available in IAEA’s TRS 483) to collect and 
validate data is essential for an accurate machine build.
Lauren WEINSTEIN (South San Francisco, USA), Matthew SKINNER
09:00 - 18:00 #17809 - Use of a Trial Setup to Assess Patient Acceptability for Mask-Based Gamma Knife Treatment.
Use of a Trial Setup to Assess Patient Acceptability for Mask-Based Gamma Knife Treatment.

Purpose: Mask-based stereotactic radiosurgery (SRS) on the Gamma Knife (GK) is facilitated by intra-fraction motion management (IFMM) and cone beam computed tomography (CBCT). The purpose of this study is to evaluate the use of a trial setup during mask simulation to determine patient eligibility and acceptability for mask-based GK-SRS.

Methods: Patients triaged to masked-based GK-SRS undergo a simulation appointment to manufacture a patient-specific mask and headrest, and acquire a reference CBCT for treatment planning. During simulation, while the mask sets on the patient, a trial setup using the IFMM device was performed by tracking a reflective marker on the patient’s nose through the stereoscopic camera. Passive tracking of the marker was performed for 10-20 minutes to assess both the magnitude of patient motion, and patient tolerability to the treatment setup position as large motions tracked via the IFMM system triggers treatment interruption during GK-SRS delivery.

 Results: Mask simulations on the GK were performed on 92 patients. Twelve patients were subsequently aborted for SRS due to various reasons following trial setup: extreme claustrophobia (2), large range of motion on IFMM due to poor performance status (2), triaged to linear accelerator treatments due to number of lesions or lesion size (6), triaged to surgical intervention (1), and patient declined GK-SRS (1). Due to collision risks identified in the pre-planning process, 8 patients were re-simulated with a new mask or foam headrest prior to GK-SRS. The IFMM threshold for GK-SRS was increased from 1.5 mm to 2.0 mm for 5 patients based on motion observed at trial setup.

Conclusions: A trial setup with the mask and IFMM system during the simulation appointment is useful in identifying patient eligibility for mask-based SRS. Identifying patient specific factors, such as claustrophobia and poor performance status, assists with patient comfort measures or triage to other more suitable treatment options.

Winnie LI, Messeret TAMEROU, David SHULTZ, Normand LAPERRIERE, Barbara-Ann MILLAR, David JAFFRAY (Toronto, CANADA), Caroline CHUNG, Catherine COOLENS
09:00 - 18:00 #17742 - Using a high frequency sampling electrometer to measure directly six time related parameters in one acquisition.
Using a high frequency sampling electrometer to measure directly six time related parameters in one acquisition.

This study is to report a novel method to measure six time related parameters: 1) timer accuracy; 2)  timer error; 3) timer linearity; 4) time between two shots; 5) time for sector passing the 4mm cone when traveling to 16 mm cone; 6) transient time that corresponding to shutter dose. This method uses a high frequency sampling electrometer to measure time directly vs the previous method that measure dose and calculate time. Current of every 0.05 sec is collected using a SunNuclear PCElectrometer in the standard GK icon QA setup over one composite acquisition of four shots of 1, 3, 10, and 20 min, respectively. The source sectors start from the sector off position, pass 4mm cone, moves to a 16 mm cone position for nominal exposure time of 1 min, then pass the 4mm cone again, and move back to the sector off position for a short period of time (time between two shots), then repeat the cycle for 3, 10, and 20 min settings, respectively. A MATLAB program was written to graph and calculate 1) timer accuracy of 59.8, 179.7, 599.5 and 1199.1 sec for nominal setting of 60, 180, 600 and 1200 sec, respectively; 2) timer error of 0.40 sec; 3) Timer linearity of R square of 1.0; 4) time between two shots of 4.9 sec; 5) time for sector passing 4mm cone of 0.4 sec; 6) transient time of 0.24 sec for the sectors to reach their full exposure positions.  This transient time is what Gamma Plan uses to account for the shutter dose.  In conclusion we have developed a semi-automatic method to measure time related parameters objectively, accurately, and efficiently by using a programmable high frequency sampling electrometer.

Tanxia QU (New York, USA), Kenneth BERNSTEIN, Douglas KONDZIOLKA