Manisha K. Ranade

Important considerations for radiochromic film dosimetry with flatbed CCD scanners and EBT GAFCHROMIC® film

In this study, we present three significant artifacts that have the potential to negatively impact the accuracy and precision of film dosimetry measurements made using GAFCHROMIC® EBT radiochromic film when read out with CCD flatbed scanners. Films were scanned using three commonly employed instruments: a Macbeth TD932 spot densitometer, an Epson Expression 1680 CCD array scanner, and a Microtek ScanMaker i900 CCD array scanner. For the two scanners we assessed the variation in optical density (OD) of GAFCHROMIC EBT film with scanning bed position, angular rotation of the film with respect to the scan line direction, and temperature inside the scanner due to repeated scanning. Scanning uniform radiochromic films demonstrated a distinct bowing effect in profiles in the direction of the CCD array with a nonuniformity of up to 17%. Profiles along a direction orthogonal to the CCD array demonstrated a 7% variation. A strong angular dependence was found in measurements made with the flatbed scanners; the effect could not be reproduced with the spot densitometer. An IMRT quality assurance film was scanned twice rotating the film 90° between the scans. For films scanned on the Epson scanner, up to 12% variation was observed in unirradiated EBT films rotated between 0° and 90°, which decreased to approximately 8% for EBT films irradiated to 300cGy. Variations of up to 80% were observed for films scanned with the Microtek scanner. The scanners were found to significantly increase the film temperature with repeated scanning. Film temperature between 18 and 33°C caused OD changes of approximately 7%. Considering these effects, we recommend adherence to a strict scanning protocol that includes: maintaining the orientation of films scanned on flatbed scanners, limiting scanning to the central portion of the scanner bed, and limiting the number of consecutive scans to minimize changes in OD caused by film heating.

Initial evaluation of commercial optical CT-based 3D gel dosimeter.

We evaluated the OCTOPUS-ONE research laser CT scanner developed and manufactured by MGS Research, Inc. (Madison, CT). The scanner is designed for imaging 3D optical density distributions in BANG gels. The scanner operates in a translate-rotate configuration with a single scanning laser beam. The rotating cylindrical gel phantom is immersed in a refractive index matching solution and positioned at the center of a square tank made of plastic and glass. A stationary polarized He-Ne laser beam (633 nm) is reflected from a mirror moving parallel to the tank wall and scans the gel. Another mirror moves synchronously along the opposite side of the tank and collects the transmitted light and sends it to a single stationary silicon photodetector. A filtered backprojection algorithm is used to reconstruct projection data in a plane. The laser-mirrors-detector assembly is mounted on a horizontal platform that moves vertically for slice selection. We have tested the mechanical and optical setup, projection centering on the axis of rotation, linearity, and spatial resolution. We found the optical detector to respond linearly to transmitted light from control samples. The spatial resolution of the scanner was determined by employing a split field resolution technique. We obtained the horizontal and vertical full widths at half maxima of the laser beam intensity profiles as 0.6 and 0.8 mm, respectively. Dose calibration tests of the gel were performed using a nine-field (2 x 2 cm2 each) dose pattern irradiated at different dose levels. Finally, we compared gel-derived 2D planar dose distribution against radiochromic film measured dose distribution for both the nine-field and a uniform 5 x 5 cm2 field of 6 MV x rays. Very similar dose distributions were observed in gel and radiochromic film except in regions of steep dose gradient and highest dose. A dose normalization of 15.6% was required between the two dosimeters due to differences in overall radiation response. After normalization, analysis using the gamma evaluation showed that the radiochromic film and gel-measured dose distributions differed by a maximum gamma of 1.3 using 5% and 1.5 mm dose difference and distance-to-agreement criteria. The optical CT scanner has great potential as a 3D dosimeter, but a few refinements and further testing are necessary before its routine clinical use.

Image registration of BANG® gel dose maps for quantitative dosimetry verification

BACKGROUND The BANG (product symbol SGEL, MGS Research Inc., Guilford, CT) polymer gel has been shown to be a valuable dosimeter for determining three-dimensional (3D) dose distributions. Because the proton relaxation rate (R2) of the gel changes as a function of absorbed dose, MR scans of the irradiated gel can be used to generate 3D dose maps. Previous work with the gel, however, has not relied on precise localization of the measured dose distribution. This has limited its quantitative use, as no precise correlation exists with the planned distribution. This paper reports on a technique for providing this correlation, thus providing a quality assurance tool that includes all of the steps of imaging, treatment planning, dose calculation, and treatment localization. METHODS AND MATERIALS The BANG gel formulation was prepared and poured into spherical flasks (15.3-cm inner diameter). A stereotactic head ring was attached to each flask. Three magnetic resonance imaging (MRI) and computed tomography (CT) compatible fiducial markers were placed on the flask, thus defining the central axial plane. A high-resolution CT scan was obtained of each flask. These images were transferred to a radiosurgery treatment-planning program, where treatment plans were developed. The gels were irradiated using our systems for stereotactic radiosurgery or fractionated stereotactic radiotherapy. The gels were MR imaged, and a relative 3D dose map was created from an R2 map of these images. The dose maps were transferred to an image-correlation program, and then fused to the treatment-planning CT scan through a rigid body match of the MRI/CT-compatible fiducial markers. The fused dose maps were imported into the treatment-planning system for quantitative comparison with the calculated treatment plans. RESULTS Calculated and measured isodose surfaces agreed to within 2 mm at the worst points within the in-plane dose distributions. This agreement is excellent, considering that the pixel resolution of the MRI dose maps is 1.56 x 1.56 mm, and the treatment-planning dose distributions were calculated on a 1-mm dose grid. All points within the dose distribution were well within the tolerances set forth for commissioning and quality assurance of stereotactic treatment-planning systems. Moreover, the quantitative evaluation presented here tests the accuracy of the entire treatment-planning and delivery process, including stereotactic frame rigidity, CT localization, CT/MR correlation, dose calculation, and radiation delivery. CONCLUSION BANG polymer gel dosimetry coupled with image correlation provides quantitative verification of the accuracy of 3D dose distributions. Such quantitative evaluation is imperative to ensure the high quality of the 3D dose distributions generated and delivered by stereotactic and other conformal irradiation systems.

Verification of step-and-shoot IMRT delivery using a fast video-based electronic portal imaging device.

We present an investigation into the use of a fast video-based electronic portal-imaging device (EPID) to study intensity modulated radiation therapy (IMRT) delivery. The aim of this study is to test the feasibility of using an EPID system to independently measure the orchestration of collimator leaf motion and beam fluence; simultaneously measuring both the delivered field fluence and shape as it exits the accelerator head during IMRT delivery. A fast EPID that consists of a terbium-doped gadolinium oxysulphide (GdO2S:Tb) scintillator coupled with an inexpensive commercial 30 frames-per-second (FPS) CCD-video recorder (16.7 ms shutter time) was employed for imaging IMRT delivery. The measurements were performed on a Varian 2100 C/D linear accelerator equipped with a 120-leaf multileaf-collimator (MLC). A characterization of the EPID was performed that included measurements of spatial resolution, linac pulse-rate dependence, linear output response, signal uniformity, and imaging artifacts. The average pixel intensity for fields imaged with the EPID was found to be linear in the delivered monitor units of static non-IMRT fields between 3x3 and 15x15 cm2. A systematic increase of the average pixel intensity was observed with increasing field size, leading to a maximum variation of 8%. Deliveries of a clinical step-and-shoot mode leaf sequence were imaged at 600 MU/min. Measurements from this IMRT delivery were compared with experimentally validated MLC controller log files and were found to agree to within 5%. An analysis of the EPID image data allowed identification of three types of errors: (1) 5 out of 35 segments were undelivered; (2) redistributing all of the delivered segment MUs; and (3) leaf movement during segment delivery. Measurements with the EPID at lower dose rates showed poor agreement with log files due to an aliasing artifact. The study was extended to use a high-speed camera (1-1000 FPS and 10 micros shutter time) with our EPID to image the same delivery to demonstrate the feasibility of imaging without aliasing artifacts. High-speed imaging was shown to be a promising direction toward validating IMRT deliveries with reasonable image resolution and noise.

A prototype quantitative film scanner for radiochromic film dosimetry

We have developed a high resolution, quantitative, two-dimensional optical film scanner for use with a commercial high sensitivity radiochromic film (RCF) for measuring single fraction external-beam radiotherapy dose distributions. The film scanner was designed to eliminate artifacts commonly observed in RCF dosimetry. The scanner employed a stationary light source and detector with a moving antireflective glass film platen attached to a high precision computerized X-Y translation stage. An ultrabright red light emitting diode (LED) with a peak output at 633 nm and full width at half maximum (FWHM) of 16 nm was selected as the scanner light source to match the RCF absorption peak. A dual detector system was created using two silicon photodiode detectors to simultaneously measure incident and transmitted light. The LED light output was focused to a submillimeter (FWHM 0.67 mm) spot size, which was determined from a scanning knife-edge technique for measuring Gaussian optical beams. Data acquisition was performed with a 16-bit A/D card in conjunction with commercial software. The linearity of the measured densities on the scanner was tested using a calibrated neutral-density step filter. Sensitometric curves and three IMRT field scans were acquired with a spatial resolution of 1 mm for both radiographic film and RCF. The results were compared with measurements taken with a commercial diode array under identical delivery conditions. The RCF was rotated by 90 deg and rescanned to study orientation effects. Comparison between the RCF and the diode array measurements using percent dose difference and distance-to-agreement criteria produced average passing rates of 99.0% using 3%/3 mm criteria and 96.7% using 2%/2 mm criteria. The same comparison between the radiographic film and diode array measurements resulted in average passing rates 96.6% and 91.6% for the above two criteria, respectively. No measurable light-scatter or interference scanner artifacts were observed. The RCF rotated by 90 deg showed no measurable orientation effect. A scan of a 15 x 15 cm2 area with 1 mm resolution required 22 min to acquire. The LED densitometer provides an accurate film dosimetry system with 1 mm or better resolution. The scanner eliminates the orientation dependence of RCF dosimetry that was previously reported with commercial flatbed scanners.

Laser microbeam CT scanning of dosimetry gels

A novel design of an optical tomographic scanner is described that can be used for 3D mapping of optical attenuation coefficient within translucent cylindrical objects with spatial resolution on the order of 100 microns. Our scanner design utilizes the cylindrical geometry of the imaged object to obtain the desired paths of the scanning light rays. A rotating mirror and a photodetector are placed at two opposite foci of the translucent cylinder that acts as a cylindrical lens. A He-Ne laser beam passes first through a focusing lens and then is reflected by the rotating mirror, so as to scan the interior of the cylinder with focused and parallel paraxial rays that are subsequently collected by the photodetector to produce the projection data, as the cylinder rotates in small angle increments between projections. Filtered backprojection is then used to reconstruct planar distributions of optical attenuation coefficient in the cylinder. Multiplanar scans are used to obtain a complete 3D tomographic reconstruction. Among other applications, the scanner can be used in radiation therapy dosimetry and quality assurance for mapping 3D radiation dose distributions in various types of tissue-equivalent gel phantoms that change their optical attenuation coefficients in proportion to the absorbed radiation dose.

A high-speed scintillation-based electronic portal imaging device to quantitatively characterize IMRT delivery

We have developed an electronic portal imaging device (EPID) employing a fast scintillator and a high-speed camera. The device is designed to accurately and independently characterize the fluence delivered by a linear accelerator during intensity modulated radiation therapy (IMRT) with either step-and-shoot or dynamic multileaf collimator (MLC) delivery. Our aim is to accurately obtain the beam shape and fluence of all segments delivered during IMRT, in order to study the nature of discrepancies between the plan and the delivered doses. A commercial high-speed camera was combined with a terbium-doped gadolinium-oxy-sulfide (Gd2O2S:Tb) scintillator to form an EPID for the unaliased capture of two-dimensional fluence distributions of each beam in an IMRT delivery. The high speed EPID was synchronized to the accelerator pulse-forming network and gated to capture every possible pulse emitted from the accelerator, with an approximate frame rate of 360 frames-per-second (fps). A 62-segment beam from a head-and-neck IMRT treatment plan requiring 68 s to deliver was recorded with our high speed EPID producing approximately 6 Gbytes of imaging data. The EPID data were compared with the MLC instruction files and the MLC controller log files. The frames were binned to provide a frame rate of 72 fps with a signal-to-noise ratio that was sufficient to resolve leaf positions and segment fluence. The fractional fluence from the log files and EPID data agreed well. An ambiguity in the motion of the MLC during beam on was resolved. The log files reported leaf motions at the end of 33 of the 42 segments, while the EPID observed leaf motions in only 7 of the 42 segments. The static IMRT segment shapes observed by the high speed EPID were in good agreement with the shapes reported in the log files. The leaf motions observed during beam-on for step-and-shoot delivery were not temporally resolved by the log files.

Imaging linear accelerator output using a high-speed scintillation based electronic portal imaging device (Hi-EPID)

We have developed a high-speed scintillation based electronic portal imaging device (Hi-EPID) to image the fluence output of a linear accelerator as it exits the head. The Hi-EPID consists of a high speed digital camera coupled with a fast terbium-doped gadolinium-oxy-sulfide (Gd2O2S:Tb) scintillator. The camera can record every single radiation pulse out of the accelerator, at up to 500 frames per second. Since the Hi-EPID allows us to simultaneously obtain both the fluence and the field shape in a delivery, it serves as an independent means of verifying treatment delivery. In this study, we used the Hi-EPID to analyze the characteristics of the fluence delivered from a commercial clinical linear accelerator (Synergy® S, Elekta). Both open field and multileaf collimator (MLC) based intensity modulated radiation therapy (IMRT) fields were evaluated. Fluence delivery irregularities were observed with the Hi-EPID device. Fluence was not constant over the duration of a segment, resulting in a varying dose rate. For the open field delivery, the fluence increased with time until it reached a plateau (in approximately 100 ms), remaining constant over the remainder of the delivery. However, in the case of a six segment IMRT field delivery, the fluence variation differed between segments. For some segments, the fluence increased in the beginning and then reached a plateau. For others, the fluence rose and fell sharply, followed by a slower rise to the plateau. The rapid fluence variation before a plateau was as high as 50% of the maximum. Though the total fluence was correctly delivered, this result is intriguing and calls for further investigation. An analysis of field shapes showed no unplanned motion of MLC during step and shoot delivery, as has been reported for another clinical linear accelerator.

SU‐FF‐T‐127: Comparison of Two Commercial Detector Arrays for IMRT Quality Assurance

Two commercially available detector arrays were compared for their use in the quality assurance of patient-specific IMRT treatment plans: one a diode-based array (MapCHECK) and the other an ion chamber-based array (MatriXX). The dependence of the response of detectors on the field size, dose rate, and radiation energy were measured and compared with reference measurements using a Farmer-type ionization chamber. The linearity of the detector response, short-term and long-term reproducibility, statistical uncertainty as a function of delivered dose, and the validity of the array calibration were also examined to understand the stability and uncertainty of the systems. No field size or SSD dependence were observed within the range of the field sizes and SSDs used in the study at both 6 MV and 18 MV photon energies. Both detector arrays showed negligible errors (< 1%) when measuring doses of more than ~8 cGy, but exhibited errors of ~3% when measuring doses on the order of 1 cGy. While the MapCHECK showed a stable short-term reproducibility to within the measurement errors, the MatriXX showed a slow but continuously increase in reading during the one-hour period (about 0.8%). The MapCHECK also showed a slightly better array sensitivity correction with all the detectors having less than 1% discrepancy and more than 90% of the detectors within 0.5% variation, whereas about 60% of the MatriXX detectors showed a less than 0.5% variation and approximately 8% exhibited a larger than 1% discrepancy. MatriXX detectors also displayed a volume-averaging effect consistent with its detector size of approximately 4.5 mm in diameter. Excellent passing rates were obtained for both detector arrays when compared with the planar dose distributions from the treatment planning system for three 6 MV IMRT fields and three 18 MV IMRT fields after the volume-averaging effect of the MatriXX was taken into account.

WE-E-BRA-06: Imaging Linear Accelerator Startup Using a High-Speed Scintillation Based Electronic Portal Imaging Device (Hi-EPID)

Purpose: Using the high‐speed scintillation based electronic portal imaging device (Hi‐EPID), to analyze the characteristics of the fluence delivered from a commercial clinical linear accelerator (Synergy® S, Elekta). To study the fluence output at short time scales. Method and Materials: The Hi‐EPID consists of a high speed digital camera coupled with a fast terbium‐doped gadolinium‐oxy‐sulfide (Gd2O2S:Tb) scintillator. The camera can record every single radiation pulse out of the accelerator, at up to 500 fps. Both open field and multileaf collimator(MLC) based intensity modulated radiation therapy(IMRT) fields were evaluated. The startup time of linac delivery was recorded. Results: Fluence delivery irregularities were observed with the Hi‐EPID device, within 250 ms. For the open field delivery, the fluence increased with time until it reached a plateau, in approximately 100 to 250 ms, and then remained constant. However, in the case of a six segment IMRT field delivery, the fluence variation differed between segments. For some segments, the fluence increased in the beginning and then reached a plateau. For others, the fluence rose and fell sharply, followed by a slower rise to the plateau. The rapid fluence variation before a plateau was as high as 60% of the maximum. A field size dependence of fluence was observed. An analysis of field shapes showed no unplanned motion of MLC was during step and shoot delivery, as has been reported for other clinical linac.Conclusion: The Hi‐EPID is capable of imaging the fluence out of a linear accelerator with high temporal resolution. The startup characteristics of the accelerator can be studied. For short time IMRT segments, fluence and dose rate varied over a significant portion of the segment. As the dose rate fluctuates during startup, its effect on IMRT delivery can be studied. This work was supported by NCI grant R01‐CA‐100636.