PyRaDiSe: A Python package for DICOM-RT-based auto-segmentation pipeline construction and DICOM-RT data conversion.

BACKGROUND AND OBJECTIVE: Despite fast evolution cycles in deep learning methodologies for medical imaging in radiotherapy, auto-segmentation solutions rarely run in clinics due to the lack of open-source frameworks feasible for processing DICOM RT Structure Sets. Besides this shortage, available open-source DICOM RT Structure Set converters rely exclusively on 2D reconstruction approaches leading to pixelated contours with potentially low acceptance by healthcare professionals. PyRaDiSe, an open-source, deep learning framework independent Python package, addresses these issues by providing a framework for building auto-segmentation solutions feasible to operate directly on DICOM data. In addition, PyRaDiSe provides profound DICOM RT Structure Set conversion and processing capabilities; thus, it applies also to auto-segmentation-related tasks, such as dataset construction for deep learning model training. METHODS: The PyRaDiSe package follows a holistic approach and provides DICOM data handling, deep learning model inference, pre-processing, and post-processing functionalities. The DICOM data handling allows for highly automated and flexible handling of DICOM image series, DICOM RT Structure Sets, and DICOM registrations, including 2D-based and 3D-based conversion from and to DICOM RT Structure Sets. For deep learning model inference, extending given skeleton classes is straightforwardly achieved, allowing for employing any deep learning framework. Furthermore, a profound set of pre-processing and post-processing routines is included that incorporate partial invertibility for restoring spatial properties, such as image origin or orientation. RESULTS: The PyRaDiSe package, characterized by its flexibility and automated routines, allows for fast deployment and prototyping, reducing efforts for auto-segmentation pipeline implementation. Furthermore, while deep learning model inference is independent of the deep learning framework, it can easily be integrated into famous deep learning frameworks such as PyTorch or Tensorflow. The developed package has successfully demonstrated its capabilities in a research project at our institution for organs-at-risk segmentation in brain tumor patients. Furthermore, PyRaDiSe has shown its conversion performance for dataset construction. CONCLUSIONS: The PyRaDiSe package closes the gap between data science and clinical radiotherapy by enabling deep learning segmentation models to be easily transferred into clinical research practice. PyRaDiSe is available on https://github.com/ubern-mia/pyradise and can be installed directly from the Python Package Index using pip install pyradise.

Mitigation of motion-induced artifacts in cone beam computed tomography using deep convolutional neural networks.

BACKGROUND: Cone beam computed tomography (CBCT) is often employed on radiation therapy treatment devices (linear accelerators) used in image-guided radiation therapy (IGRT). For each treatment session, it is necessary to obtain the image of the day in order to accurately position the patient and to enable adaptive treatment capabilities including auto-segmentation and dose calculation. Reconstructed CBCT images often suffer from artifacts, in particular those induced by patient motion. Deep-learning based approaches promise ways to mitigate such artifacts. PURPOSE: We propose a novel deep-learning based approach with the goal to reduce motion induced artifacts in CBCT images and improve image quality. It is based on supervised learning and includes neural network architectures employed as pre- and/or post-processing steps during CBCT reconstruction. METHODS: Our approach is based on deep convolutional neural networks which complement the standard CBCT reconstruction, which is performed either with the analytical Feldkamp-Davis-Kress (FDK) method, or with an iterative algebraic reconstruction technique (SART-TV). The neural networks, which are based on refined U-net architectures, are trained end-to-end in a supervised learning setup. Labeled training data are obtained by means of a motion simulation, which uses the two extreme phases of 4D CT scans, their deformation vector fields, as well as time-dependent amplitude signals as input. The trained networks are validated against ground truth using quantitative metrics, as well as by using real patient CBCT scans for a qualitative evaluation by clinical experts. RESULTS: The presented novel approach is able to generalize to unseen data and yields significant reductions in motion induced artifacts as well as improvements in image quality compared with existing state-of-the-art CBCT reconstruction algorithms (up to +6.3 dB and +0.19 improvements in peak signal-to-noise ratio, PSNR, and structural similarity index measure, SSIM, respectively), as evidenced by validation with an unseen test dataset, and confirmed by a clinical evaluation on real patient scans (up to 74% preference for motion artifact reduction over standard reconstruction). CONCLUSIONS: For the first time, it is demonstrated, also by means of clinical evaluation, that inserting deep neural networks as pre- and post-processing plugins in the existing 3D CBCT reconstruction and trained end-to-end yield significant improvements in image quality and reduction of motion artifacts.

Tracking latency in image-based dynamic MLC tracking with direct image access.

PURPOSE: Target tracking is a promising method for motion compensation in radiotherapy. For image-based dynamic multileaf collimator (DMLC) tracking, latency has been shown to be the main contributor to geometrical errors in tracking of respiratory motion, specifically due to slow transfer of image data from the image acquisition system to the tracking system via image file storage on a hard disk. The purpose of the current study was to integrate direct image access with a DMLC tracking system and to quantify the tracking latency of the integrated system for both kV and MV image-based tracking. METHOD: A DMLC tracking system integrated with a linear accelerator was used for tracking of a motion phantom with an embedded tungsten marker. Real-time target localization was based on x-ray images acquired either with a portal imager or a kV imager mounted orthogonal to the treatment beam. Images were processed directly without intermediate disk access. Continuous portal images and system log files were stored during treatment delivery for detailed offline analysis of the tracking latency. RESULTS: The mean tracking system latency for kV and MV image-based tracking as function of the imaging interval ΔT(image) increased linearly with ΔT(image) as 148 ms + 0.58 * ΔT(image) (kV) and 162 ms + 1.1 * ΔT(image) (MV). The latency contribution from image acquisition and image transfer for kV image-based tracking was independent on ΔT(image) at 103 ± 14 ms. For MV-based tracking, it increased with ΔT(image) as 124 ms + 0.44 * ΔT(image). For ΔT(image) = 200 ms (5 Hz imaging), the total latency was reduced from 550 ms to 264 ms for kV image-based tracking and from 500 ms to 382 ms for MV image-based tracking as compared to the previously used indirect image transfer via image file storage on a hard disk. CONCLUSION: kV and MV image-based DMLC tracking was successfully integrated with direct image access. It resulted in substantial tracking latency reductions compared with image-based tracking without direct image access.

The predictive value of segmentation metrics on dosimetry in organs at risk of the brain.

BACKGROUND: Fully automatic medical image segmentation has been a long pursuit in radiotherapy (RT). Recent developments involving deep learning show promising results yielding consistent and time efficient contours. In order to train and validate these systems, several geometric based metrics, such as Dice Similarity Coefficient (DSC), Hausdorff, and other related metrics are currently the standard in automated medical image segmentation challenges. However, the relevance of these metrics in RT is questionable. The quality of automated segmentation results needs to reflect clinical relevant treatment outcomes, such as dosimetry and related tumor control and toxicity. In this study, we present results investigating the correlation between popular geometric segmentation metrics and dose parameters for Organs-At-Risk (OAR) in brain tumor patients, and investigate properties that might be predictive for dose changes in brain radiotherapy. METHODS: A retrospective database of glioblastoma multiforme patients was stratified for planning difficulty, from which 12 cases were selected and reference sets of OARs and radiation targets were defined. In order to assess the relation between segmentation quality -as measured by standard segmentation assessment metrics- and quality of RT plans, clinically realistic, yet alternative contours for each OAR of the selected cases were obtained through three methods: (i) Manual contours by two additional human raters. (ii) Realistic manual manipulations of reference contours. (iii) Through deep learning based segmentation results. On the reference structure set a reference plan was generated that was re-optimized for each corresponding alternative contour set. The correlation between segmentation metrics, and dosimetric changes was obtained and analyzed for each OAR, by means of the mean dose and maximum dose to 1% of the volume (Dmax 1%). Furthermore, we conducted specific experiments to investigate the dosimetric effect of alternative OAR contours with respect to the proximity to the target, size, particular shape and relative location to the target. RESULTS: We found a low correlation between the DSC, reflecting the alternative OAR contours, and dosimetric changes. The Pearson correlation coefficient between the mean OAR dose effect and the Dice was -0.11. For Dmax 1%, we found a correlation of -0.13. Similar low correlations were found for 22 other segmentation metrics. The organ based analysis showed that there is a better correlation for the larger OARs (i.e. brainstem and eyes) as for the smaller OARs (i.e. optic nerves and chiasm). Furthermore, we found that proximity to the target does not make contour variations more susceptible to the dose effect. However, the direction of the contour variation with respect to the relative location of the target seems to have a strong correlation with the dose effect. CONCLUSIONS: This study shows a low correlation between segmentation metrics and dosimetric changes for OARs in brain tumor patients. Results suggest that the current metrics for image segmentation in RT, as well as deep learning systems employing such metrics, need to be revisited towards clinically oriented metrics that better reflect how segmentation quality affects dose distribution and related tumor control and toxicity.

Deep Learning model for markerless tracking in spinal SBRT.

Stereotactic Body Radiation Therapy (SBRT), alternatively termed Stereotactic ABlative Radiotherapy (SABR) or Stereotactic RadioSurgery (SRS), delivers high dose with a sub-millimeter accuracy. It requires meticulous precautions on positioning, as sharp dose gradients near critical neighboring structures (e.g. the spinal cord for spinal tumor treatment) are an important clinical objective to avoid complications such as radiation myelopathy, compression fractures, or radiculopathy. To allow for dose escalation within the target without compromising the dose to critical structures, proper immobilization needs to be combined with (internal) motion monitoring. Metallic fiducials, as applied in prostate, liver or pancreas treatments, are not suitable in clinical practice for spine SBRT. However, the latest advances in Deep Learning (DL) allow for fast localization of the vertebrae as landmarks. Acquiring projection images during treatment delivery allows for instant 2D position verification as well as sequential (delayed) 3D position verification when incorporated in a Digital TomoSynthesis (DTS) or Cone Beam Computed Tomography (CBCT). Upgrading to an instant 3D position verification system could be envisioned with a stereoscopic kilovoltage (kV) imaging setup. This paper describes a fast DL landmark detection model for vertebra (trained in-house) and evaluates its accuracy to detect 2D motion of the vertebrae with the help of projection images acquired during treatment. The introduced motion consists of both translational and rotational variations, which are detected by the DL model with a sub-millimeter accuracy.

Introducing gel dosimetry in a clinical environment: customization of polymer gel composition and magnetic resonance imaging parameters used for 3D dose verifications in radiosurgery and intensity modulated radiotherapy.

Radiation sensitive gels have been used as dosimeters for clinical dose verification of different radiation therapy modalities. However, the use of gels is not widespread, because careful techniques are required to achieve the dose precision and accuracy aimed for in clinical dose verification. Here, the introduction of gel dosimetry in a clinical environment is described, including the whole chain of customizations and preparations required to introduce magnetic resonance (MR) based gel dosimetry into clinical routine. In order to standardize gel dosimetry in dose verifications for radiosurgery and intensity modulated radiotherapy (IMRT), we focused on both the customization of the gel composition and of the MR imaging parameters to increase its precision. The relative amount of the components of the normoxic, methacrylic acid based gel (MAGIC) was changed to obtain linear and steep dose response relationships. MR imaging parameters were customized for the different dose ranges used in order to lower the relative standard deviation of the measured transversal relaxation rate (R2). An optimization parameter was introduced to quantify the change in the relative standard deviation of R2 (sigma(R2,rel)) taking the increase in MR time into account. A 9% methacrylic acid gel customized for radiosurgery was found to give a linear dose response up to 40 Gy with a slope of 0.94 Gy(-1) s(-1), while a 6% methacrylic acid gel customized for IMRT had a linear range up to 3 Gy with a slope of 1.86 Gy(-1) s(-1). With the help of an introduced optimization parameter, the mean sigma(R2,rel) was improved by 13% for high doses and by 55% for low doses, without increasing MR time to unacceptable values. A mean dose resolution of less than 0.13 Gy has been achieved with the gel and imaging parameters customized for IMRT and a dose resolution from 0.97 Gy (at 5 Gy) to 2.15 Gy (at 40 Gy) for the radiosurgery dose range. The comparisons of calculated and measured relative 3D dose distributions performed for radiosurgery and IMRT showed an acceptable overall correlation. The gamma criterion for the radiosurgery verification with a voxel size of 1.5 x 1.5 x 1.5 mm3 was passed by 96.8% of the voxels (1.5 mm distance, 8% in dose). For the IMRT verification using a voxel size of 1.25 x 1.25 x 5 mm3 the gamma criterion was passed by 50.3% of the voxels (3 mm distance, 3% dose uncertainty). Using dedicated data analysis and visualization software, MR based normoxic gel dosimetry was found to be a valuable tool for clinically based dose verification, provided that customized gel compositions and MR imaging parameters are used. While high dose precision was achieved, further work is required to achieve clinically acceptable dose accuracy.

Verifying tumor position during stereotactic body radiation therapy delivery using (limited-arc) cone beam computed tomography imaging.

BACKGROUND AND PURPOSE: Proof of tumor position during stereotactic body radiotherapy (SBRT) delivery is desirable. We investigated if cone-beam CT (CBCT) scans reconstructed from (collimated) fluoroscopic kV images acquired during irradiation could show the dominant tumor position. MATERIALS AND METHODS: Full-arc CBCT scans were reconstructed using FDK filtered back projection from 38kV fluoroscopy datasets (16 patients) continuously acquired during volumetric modulated spine SBRT. CBCT-CT match values were compared to the average spine offset values found using template matching+triangulation of the individual kV images. Multiple limited-arc CBCTs were reconstructed from fluoroscopic images acquired during lung SBRT of an anthropomorphic thorax phantom using 20-180° arc lengths and for 3 breath-hold lung SBRT patients. RESULTS: Differences between 3D CBCT-CT match results and average spine offsets found using template matching+triangulation were 0.1±0.1mm for all directions (range: 0.0-0.5mm). For limited-arc CBCTs of the thorax phantom, the automatic 3D CBCT-CT match results for arc lengths of 80-180° were ≤1mm. 20° CBCT reconstruction still allowed for positional verification in 2D. CONCLUSIONS: (Limited-arc) CBCT reconstructions of kV images acquired during irradiation can identify the dominant position of the tumor during treatment delivery.

Detection of anatomical changes in lung cancer patients with 2D time-integrated, 2D time-resolved and 3D time-integrated portal dosimetry: a simulation study.

The aim of this work is to assess the performance of 2D time-integrated (2D-TI), 2D time-resolved (2D-TR) and 3D time-integrated (3D-TI) portal dosimetry in detecting dose discrepancies between the planned and (simulated) delivered dose caused by simulated changes in the anatomy of lung cancer patients. For six lung cancer patients, tumor shift, tumor regression and pleural effusion are simulated by modifying their CT images. Based on the modified CT images, time-integrated (TI) and time-resolved (TR) portal dose images (PDIs) are simulated and 3D-TI doses are calculated. The modified and original PDIs and 3D doses are compared by a gamma analysis with various gamma criteria. Furthermore, the difference in the D 95% (ΔD 95%) of the GTV is calculated and used as a gold standard. The correlation between the gamma fail rate and the ΔD 95% is investigated, as well the sensitivity and specificity of all combinations of portal dosimetry method, gamma criteria and gamma fail rate threshold. On the individual patient level, there is a correlation between the gamma fail rate and the ΔD 95%, which cannot be found at the group level. The sensitivity and specificity analysis showed that there is not one combination of portal dosimetry method, gamma criteria and gamma fail rate threshold that can detect all simulated anatomical changes. This work shows that it will be more beneficial to relate portal dosimetry and DVH analysis on the patient level, rather than trying to quantify a relationship for a group of patients. With regards to optimizing sensitivity and specificity, different combinations of portal dosimetry method, gamma criteria and gamma fail rate should be used to optimally detect certain types of anatomical changes.

Gamma Knife surgery after fractionated radiotherapy for acromegaly.

OBJECT: Acromegaly that has not been cured by microsurgery is usually treated with fractionated radiotherapy; however, it is not possible to repeat such a treatment with effective radiation doses if it should fail. The authors pose the question: Can stereotactic radiosurgery be used as an effective, alternative method for retreatment by irradiation? METHODS: A retrospective study of 12 patients was performed to compare patients treated with Gamma Knife surgery (GKS) after initial, failed radiotherapy and 37 patients treated with GKS only. The mean dose for the initial fractionated radiotherapy was 44.6 Gy (range 40-54 Gy). The mean maximum GKS dose was 45.1 Gy (range 27-50 Gy) in the pretreated group and 49.5 Gy (range 25-70 Gy) in the group undergoing GKS alone. The mean interval between the two treatments was 10.6 years (range 3-20.6 years). The age-related insulin-like growth factor-I (IGF-I), assessed at 3-month intervals, was the main follow-up parameter. An IGF-I normalization rate of more than 80% was achieved in both patient groups; however, the latency of endocrinological normalization was longer in the patients who had undergone failed fractionated radiotherapy (median time to cure 35.4 months compared with 13.5 months). CONCLUSIONS: Treatment with GKS is successful in patients with acromegaly even after failed fractionated radiotherapy; GKS represents a therapeutic tool in patients with no therapeutic options life-long octreotide. It must be noted that the incidence of neurological complications is higher (p < 0.01, 2 x 2 crosstab). The remaining dose fraction after previous fractionated radiotherapy appears to be approximately 50%. Maintenance of other endocrinological functions may be better after GKS alone; however, the difference is not significant.

[Application of normoxic polymer gels in 3D-dosimetry for radiosurgery]. / Einsatz normoxischer Polymergele in der 3D-Dosimetrie für die Radiochirurgie.

The aim of this study was to describe the manufacture of normoxic polymer gels, to characterize their dose response relationship, to optimize MR imaging parameters in order to minimize the standard deviation in the measured dose and to use the gel in a dose verification experiment in radiosurgery. The normoxic polymer gel used is simple to manufacture under normal atmospheric conditions and is characterized by a linear dose relationship up to 40 Gy. MR imaging was performed using 2-dimensional (20) single spin echo pulse sequences with two different echo times. The imaging parameters were optimized in order to minimize the standard deviation of the measured transversal relaxation rate R2 and to achieve a geometrical resolution of 1.5 mm. Comparisons of calculated and measured relative 3D dose distributions using a multi isocentric irradiation with Gamma Knife B showed a good overall agreement of both the isodose levels and the differential and cumulative dose volume histograms. The standard deviation in the measured dose was approximately 9% at 30 Gy. The evaluation according to the gamma criterion showed that 96% of the dose voxels remained within a spatial uncertainty of 1.5 mm and a dose uncertainty of 8%.

Analyzing 3-tesla magnetic resonance imaging units for implementation in radiosurgery.

OBJECT: The limiting factor affecting accuracy during gamma knife surgery is image quality. The new generation of magnetic resonance (MR) imaging units with field strength up to 3 teslas promise superior image quality for anatomical resolution and contrast. There are, however, questions about chemical shifts or susceptibility effects, which are the subject of this paper. METHODS: The 3-tesla MR imaging unit (Siemens Trio) was analyzed and compared with a 1-tesla unit (Siemens Magnetom Expert) and to a 1.5-tesla unit (Philips Gyroscan). Evaluation of the magnitude of error was performed within transverse slices in two orientations (axial/coronal) by using a cylindrical phantom with an embedded grid. Deviations were determined for 21 targets in a slab phantom with known geometrical positions within the stereotactic frame. Distortions caused by chemical shift and/or susceptibility effects were analyzed in a head phantom. Inhouse software was used for data analyses. The mean deviation was less than 0.3 mm in axial and less than 0.4 mm in coronal orientations. For the known targets the maximum deviation was 1.16 mm. By optimizing these parameters in the protocol these inaccuracies could be reduced to less than 1.1 mm. Due to inhomogeneities a shift in the z direction of up to 1.5 mm was observed for a dataset, which was shown to be compressed by 1.2 mm. CONCLUSIONS: The 3-tesla imaging unit showed superior anatomical contrast and resolution in comparison with the established 1-tesla and 1.5-tesla units; however, due to the high field strength the field within the head coil is very sensitive to inhomogeneities and therefore 3-tesla imaging data will have be handled with care.

Quantifying the degree of conformity in radiosurgery treatment planning.

PURPOSE: To compare different parameters used to quantify the quality of a treatment plan and to evaluate the dose conformity and coverage clinically achieved using gamma knife radiosurgery. METHODS AND MATERIALS: Various existing parameters for coverage and conformity are reviewed. Additionally, a modified conformity index (CI) has been defined as the ratio of the volume within the target irradiated to at least the prescription isodose over the total volume enclosed by the prescription isodose. These parameters are calculated for all the 551 evaluable patient treatment plans. RESULTS: The median CI for all targets is 0.75, with a median target coverage of 94.6%. Regardless of the conformity parameter chosen, the conformity is seen to vary depending on the type of tumor and its location, reflecting the treatment planning philosophy. For tumors with volumes smaller than about 1 cm(3), the conformity parameter is also seen to be dependent on the target volume. CONCLUSION: With gamma knife radiosurgery, it is possible to achieve highly conformal dose distributions. A single parameter for the quantification of a plan, though desirable, is not realistic, because of the competing components of high dose to the target and low dose to normal tissue. Thus, we propose the use of the CI, together with the target volume coverage.

High precision film dosimetry with GAFCHROMIC films for quality assurance especially when using small fields.

Treatment units for radiosurgery, brachytherapy, implementation of seeds, and IMRT generate small high dose regions together with steep dose gradients of up to 30%-50% per mm. Such devices are used to treat small complex-shaped lesions, often located close to critical structures, by superimposing several single high dose regions. In order to test and verify these treatment techniques, to perform quality assurance tasks and to simulate treatment conditions as well as to collect input data for treatment planning, a GAFCHROMIC film based dosimetry system for measuring two-dimensional (2-D) and three-dimensional (3-D) dose distributions was developed. The nearly tissue-equivalent radiochromic GAFCHROMIC film was used to measure dose distributions. A drum scanner was investigated and modified. The spectral emission of the light source and the filters together with the efficiency of the CCD filters for the red color were matched and balanced with the absorption spectra of the film. Models based on refined studies have been developed to characterize theoretically the physics of film exposure and to calibrate the film. Mathematical descriptions are given to calculate optical densities from spectral data. The effect of darkening has been investigated and is described with a mathematical model. The influence of the scan temperature has been observed and described. In order to cope with the problem of individual film inhomogeneities, a double irradiation technique is introduced and implemented that yields dose accuracies as good as 2%-3%. Special software routines have been implemented for evaluating and handling the film data.

Precision dosimetry for narrow photon beams used in radiosurgery-determination of Gamma Knife output factors.

Treatment units for radiosurgery, like Leksell Gamma Knife and adapted, or dedicated, linear accelerators use small circular beams of ionizing radiation down to 4 mm in diameter at the isocenter. By cross-firing, these beams generate a high dose region at the isocenter together with steep dose gradients of up to 30% per mm. These units are used to treat small complex shaped lesions, often located close to critical structures within the brain, by superimposing several single high dose regions. In order to commission such treatment units for stereotactic irradiations, to carry out quality assurance and to simulate treatment conditions, as well as to collect input data for treatment planning, a precise dosimetric system is necessary. Commercially available radiation dosimeters only partially meet the requirements for narrow photon beams and small field sizes as used in stereotactic treatment modalities. The aim of this study was the experimental determination of the output factors for the field defining collimators used in Gamma Knife radiosurgery, in particular for the smallest, the 4 mm collimator helmet. For output factor measurements a pin point air ionization chamber, a liquid ionization chamber, a diode detector, a diamond detector, TLD microcubes and microrods, alanine pellets, and radiochromic films were used. In total, more than 1000 measurements were performed with these different detection systems, at the sites in Munich and Zurich. Our results show a resultant output factor for the 4 mm collimator helmet of 0.8741 +/- 0.0202, which is in good agreement with recently published results and demonstrates the feasibility of such measurements. The measured output factors for the 8 mm and 14 mm collimator helmets are 0.9578 +/- 0.0057 and 0.9870 +/- 0.0086, respectively.

Three-dimensional dose verification using BANG gel: a clinical example.

OBJECT: The authors sought to demonstrate the possible value of three-dimensional dose verification by using gel dosimetry. METHODS: In this study, commercially available BANG-25 Gy gel was used. This polymer gel is tissue equivalent and the relaxation rate (R2) measured using magnetic resonance (MR) imaging is proportional to the absorbed dose in the gel. A cylindrical container filled with BANG was mounted within an anthropomorphic head phantom and was handled using the same process as would be used for a patient undergoing gamma knife radiosurgery (GKS). An irregular target outline was constructed and a dose plan was created consisting of seven shots, three using the 8-mm and four using the 4-mm collimator helmet. The maximum dose specified was 25 Gy. A combination of several single spin-echo MR imaging sequences with different echo times was used to calculate the R2. The geometric resolution of the MR images was approximately 1 mm3. To compare the measured dose distribution with the calculated one, isodoses were overlaid in three orthogonal planes by using specially designed analysis software. CONCLUSIONS: Comparisons of the measured and calculated relative dose distributions showed good overall agreement, with differences of less than 3 mm between measured and calculated isodoses. High resolution BANG gel dosimetry for GKS can be useful for the verification of clinical treatment plans, especially when multiple shots are involved. Further verifications will be done using additional imaging parameters and absolute dose calibrations to improve the method.

Time dependent pre-treatment EPID dosimetry for standard and FFF VMAT.

Methods to calibrate Megavoltage electronic portal imaging devices (EPIDs) for dosimetry have been previously documented for dynamic treatments such as intensity modulated radiotherapy (IMRT) using flattened beams and typically using integrated fields. While these methods verify the accumulated field shape and dose, the dose rate and differential fields remain unverified. The aim of this work is to provide an accurate calibration model for time dependent pre-treatment dose verification using amorphous silicon (a-Si) EPIDs in volumetric modulated arc therapy (VMAT) for both flattened and flattening filter free (FFF) beams. A general calibration model was created using a Varian TrueBeam accelerator, equipped with an aS1000 EPID, for each photon spectrum 6 MV, 10 MV, 6 MV-FFF, 10 MV-FFF. As planned VMAT treatments use control points (CPs) for optimization, measured images are separated into corresponding time intervals for direct comparison with predictions. The accuracy of the calibration model was determined for a range of treatment conditions. Measured and predicted CP dose images were compared using a time dependent gamma evaluation using criteria (3%, 3 mm, 0.5 sec). Time dependent pre-treatment dose verification is possible without an additional measurement device or phantom, using the on-board EPID. Sufficient data is present in trajectory log files and EPID frame headers to reliably synchronize and resample portal images. For the VMAT plans tested, significantly more deviation is observed when analysed in a time dependent manner for FFF and non-FFF plans than when analysed using only the integrated field. We show EPID-based pre-treatment dose verification can be performed on a CP basis for VMAT plans. This model can measure pre-treatment doses for both flattened and unflattened beams in a time dependent manner which highlights deviations that are missed in integrated field verifications.

Dosimetric properties of an amorphous silicon EPID for verification of modulated electron radiotherapy.

PURPOSE: To investigate the dosimetric properties of an electronic portal imaging device (EPID) for electron beam detection and to evaluate its potential for quality assurance (QA) of modulated electron radiotherapy (MERT). METHODS: A commercially available EPID was used to detect electron beams shaped by a photon multileaf collimator (MLC) at a source-surface distance of 70 cm. The fundamental dosimetric properties such as reproducibility, dose linearity, field size response, energy response, and saturation were investigated for electron beams. A new method to acquire the flood-field for the EPID calibration was tested. For validation purpose, profiles of open fields and various MLC fields (square and irregular) were measured with a diode in water and compared to the EPID measurements. Finally, in order to use the EPID for QA of MERT delivery, a method was developed to reconstruct EPID two-dimensional (2D) dose distributions in a water-equivalent depth of 1.5 cm. Comparisons were performed with film measurement for static and dynamic monoenergy fields as well as for multienergy fields composed by several segments of different electron energies. RESULTS: The advantageous EPID dosimetric properties already known for photons as reproducibility, linearity with dose, and dose rate were found to be identical for electron detection. The flood-field calibration method was proven to be effective and the EPID was capable to accurately reproduce the dose measured in water at 1.0 cm depth for 6 MeV, 1.3 cm for 9 MeV, and 1.5 cm for 12, 15, and 18 MeV. The deviations between the output factors measured with EPID and in water at these depths were within ±1.2% for all the energies with a mean deviation of 0.1%. The average gamma pass rate (criteria: 1.5%, 1.5 mm) for profile comparison between EPID and measurements in water was better than 99% for all the energies considered in this study. When comparing the reconstructed EPID 2D dose distributions at 1.5 cm depth to film measurements, the gamma pass rate (criteria: 2%, 2 mm) was better than 97% for all the tested cases. CONCLUSIONS: This study demonstrates the high potential of the EPID for electron dosimetry, and in particular, confirms the possibility to use it as an efficient verification tool for MERT delivery.