Kasper L. Pasma

Dosimetric verification of intensity modulated beams produced with dynamic multileaf collimation using an electronic portal imaging device.

Dose distributions can often be significantly improved by modulating the two-dimensional intensity profile of the individual x-ray beams. One technique for delivering intensity modulated beams is dynamic multileaf collimation (DMLC). However, DMLC is complex and requires extensive quality assurance. In this paper a new method is presented for a pretreatment dosimetric verification of these intensity modulated beams utilizing a charge-coupled devicecamera based fluoroscopic electronic portal imaging device(EPID). In the absence of the patient, EPIDimages are acquired for all beams produced with DMLC. These images are then converted into two-dimensional dose distributions and compared with the calculated dose distributions. The calculations are performed with a pencil beam algorithm as implemented in a commercially available treatment planning system using the same absolute beam fluence profiles as used for calculation of the patient dose distribution. The method allows an overall verification of (i) the leaf trajectory calculation (including the models to incorporate collimator scatter and leaf transmission), (ii) the correct transfer of the leaf sequencing file to the treatment machine, and (iii) the mechanical and dosimetrical performance of the treatment unit. The method was tested for intensity modulated 10 and 25 MV photon beams; both model cases and real clinical cases were studied. Dose profiles measured with the EPID were also compared with ionization chamber measurements. In all cases both predictions and EPID measurements and EPID and ionization chamber measurements agreed within 2% (1σ). The study has demonstrated that the proposed method allows fast and accurate pretreatment verification of DMLC.

Portal dose image (PDI) prediction for dosimetric treatment verification in radiotherapy. I. An algorithm for open beams

A method is presented for calculation of transmission functions for high energy photon beams through patients. These functions are being used in our clinic for prediction of portal dose images (PDIs) which are compared with PDIs measured with an electronic portal imaging device (EPID). The calculations are based on the planning CT-scan of the patient and on the irradiation geometry as determined in the treatment planning process. For each beam quality, the required input data for the algorithm for transmission prediction are derived from a limited number of measured beam data. The method has been tested for a PDI-plane at 160 cm from the focus, in agreement with the fixed focus-to-detector distance of our fluoroscopic EPIDs. For 6, 23 and 25 MV photon beams good agreement (approximately 1%) has been found between calculated and measured transmissions through anthropomorphic phantoms.

In vivo dosimetry for prostate cancer patients using an electronic portal imaging device (EPID); demonstration of internal organ motion

PURPOSE To investigate the use of a commercially available video-based EPID for in vivo dosimetry during treatment of prostate cancer patients. METHODS For 10 prostate cancer patients, the inter-fraction variation within measured portal dose images (PDIs) was assessed and measured PDIs were compared with corresponding predicted PDIs based on the planning CT scan of the patient. RESULTS For the lateral fields, the average standard deviation in the measured on-axis portal doses during the course of a treatment was 0.9%; for the anterior fields this standard deviation was 2.2%. The difference between the average on-axis measured portal dose and the predicted portal dose was 0.3+/-2.1% (1 SD) for the lateral fields and 0.7+/-3.4% (1 SD) for the anterior fields. Off-axis differences between measured and predicted portal doses were regularly much larger (up to 15%) and were caused by frequently occurring gas pockets inside the rectum of the patients during treatment or during acquisition of the planning CT scan. The detected gas pockets did sometimes extend into the gross tumour volume (GTV) area as outlined in the planning CT scans, implying a shift of the anterior rectum wall and prostate in the anterior direction (internal organ motion). CONCLUSIONS The developed procedures for measurement and prediction of PDIs allow accurate dosimetric quality control of the treatment of prostate cancer patients. Comparing measured PDIs with predicted PDIs can reveal internal organ motion.

Fast and accurate leaf verification for dynamic multileaf collimation using an electronic portal imaging device.

A prerequisite for accurate dose delivery of IMRT profiles produced with dynamic multileaf collimation (DMLC) is highly accurate leaf positioning. In our institution, leaf verification for DMLC was initially done with film and ionization chamber. To overcome the limitations of these methods, a fast, accurate and two-dimensional method for daily leaf verification, using our CCD-camera based electronic portal imaging device (EPID), has been developed. This method is based on a flat field produced with a 0.5 cm wide sliding gap for each leaf pair. Deviations in gap widths are detected as deviations in gray scale value profiles derived from the EPID images, and not by directly assessing leaf positions in the images. Dedicated software was developed to reduce the noise level in the low signal images produced with the narrow gaps. The accuracy of this quality assurance procedure was tested by introducing known leaf position errors. It was shown that errors in leaf gap as small as 0.01-0.02 cm could be detected, which is certainly adequate to guarantee accurate dose delivery of DMLC treatments, even for strongly modulated beam profiles. Using this method, it was demonstrated that both short and long term reproducibility in leaf positioning were within 0.01 cm (1sigma) for all gantry angles, and that the effect of gravity was negligible.

Dosimetric verification of x-ray fields with steep dose gradients using an electronic portal imaging device.

Regions with steep dose gradients are often encountered in clinical x-ray beams, especially with the growing use of intensity modulated radiotherapy (IMRT). Such regions are present both at field edges and, for IMRT, in the vicinity of the projection of sensitive anatomical structures in the treatment field. Dose measurements in these regions are often difficult and labour intensive, while dose prediction may be inaccurate. A dedicated algorithm developed in our institution for conversion of pixel values, measured with a charged coupled device camera based fluoroscopic electronic portal imaging device (EPID), into absolute absorbed doses at the EPID plane has an accuracy of 1-2% for flat and smoothly modulated fields. However, in the current algorithm there is no mechanism to correct for the (short-range) differences in lateral electron transport between water and the metal plate with the fluorescent layer in the EPID. Moreover, lateral optical photon transport in the fluorescent layer is not taken into account. This results in large deviations (>10%) in the penumbra region of these fields. We have investigated the differences between dose profiles measured in water and with the EPID for small heavily peaked fields. A convolution kernel has been developed to empirically describe these differences. After applying the derived kernel to raw EPID images, a general agreement within 2% was obtained with the water measurements in the central region of the fields, and within 0.03 cm in the penumbra region. These results indicate that the EPID is well suited for accurate dosimetric verification of steep gradient x-ray fields.

Verification of compensator thicknesses using a fluoroscopic electronic portal imaging device

A method is presented for verification of compensator thicknesses using a fluoroscopic electronic portal imaging device (EPID). The method is based on the measured transmission through the compensator, defined by the ratio of the portal dose with the compensator in the beam and the portal dose without the compensator in the beam. The transmission is determined with the EPID by dividing two images, acquired with and without compensator inserted, which are only corrected for the nonlinear response of the fluoroscopic system. The transmission has a primary and a scatter component. The primary component is derived from the measured transmission by subtracting the predicted scatter component. The primary component for each point is only related to the radiological thickness of the compensator along the ray line between the focus and that point. Compensator thicknesses are derived from the primary components taking into account off-axis variations in beam quality. The developed method has been tested for various compensators made of a granulate of stainless steel. The compensator thicknesses could be determined with an accuracy of 0.5 mm (1 s.d.), corresponding to a change in the transmitted dose of about 1% for a 10 MV beam. The method is fast, accurate, and insensitive to long-term output and beam profile fluctuations of the linear accelerator.

Dosimetry with a fluoroscopic electronic portal imaging device

textabstractechniques for dosimetric verification of radiotherapy treatments using a CCD camera based fluoroscopic electronic portal imaging device (EPID) are described. The dosimetric characteristics of the EPID were investigated and a method was developed to derive portal dose images (PDIs) from measured EPID images. EPID and ionization chamber measurements agreed to within 1% (1 s). Subsequently, an algorithm was developed to predict these PDIs using the planning CT data of the patient and the irradiation geometry as determined in the treatment planning process. Furthermore, a method was developed to derive the on-axis patient dose from an EPID measurement, which was then compared with the intended dose. The method allows the discrimination of errors that are due to changes in patient anatomy and errors due to a deviating cGy/MU-value. For 115 prostate cancer patients the differences between the average on-axis measured portal dose and the predicted portal dose for the three open beams were small: -0.3±2.3% (1 s). However, large (up to 15%) off-axis differences between measured and predicted PDIs were found, which were caused by frequently occurring gas pockets inside the rectum of the patients during treatment or during acquisition of the planning CT scan. The detected gas pockets did sometimes extend into the tumor volume area as outlined in the CT scan, implying internal organ motion. Finally, methods were developed for pretreatment verification of intensity modulated fields produced with compensators or dynamic multileaf collimation (DMLC). EPID measurements of dose profiles generated with DMLC agreed within 2% (1 s) with predictions and ionization chamber measurements.

Using routine portal images to detect internal organ motion in prostate cancer patients; a simulation

Geometrical uncertainties in external radiotherapy consist for a large part of set-up errors and internal organ motion. Set-up deviations can be derived from portal images acquired during the treatment. The gray shades in those images reflect the irradiated radiological thicknesses and bony structures are therefore highly visible. Determining the position of the bony anatomy relative to the reference situation yields the set-up deviation. In routine portal images of pelvic fields, the most visible objects besides bony structures are gas pockets in the rectum. These gas pockets might indicate the position of the ventral rectum wall, which is expected to have a good correlation with the prostate position because of their close proximity. In Fig. 1 the idea behind this correlation is visualized. Kroonwijk et al already pointed out that gas pockets in portal images can reveal internal organ motion.[1] They did however not specify how this could be used in practice. In this abstract, a method for using the imaged gas pockets for quantitative determination of the rectum wall and prostate position is proposed. Compared to other methods which make use of e.g. radio-opaque markers or multiple CT data, this method is non-invasive and little extra work is required; portal images are already routinely made for set-up verification in our institute.

115 Dosimetric verification of intensity modulated fields produced with dynamic multileaf collimation using an electronic portal imaging device

Dose distributions can often be significantly improved by modulating the two-dimensional intensity profile of the individual x-ray beams. One technique for delivering intensity modulated beams is dynamic multileaf collimation (DMLC). However, DMLC is complex and requires extensive quality assurance. In this paper a new method is presented for a pretreatment dosimetric verification of these intensity modulated beams utilizing a charge-coupled device camera based fluoroscopic electronic portal imaging device (EPID). In the absence of the patient, EPID images are acquired for all beams produced with DMLC. These images are then converted into two-dimensional dose distributions and compared with the calculated dose distributions. The calculations are performed with a pencil beam algorithm as implemented in a commercially available treatment planning system using the same absolute beam fluence profiles as used for calculation of the patient dose distribution. The method allows an overall verification of (i) the leaf trajectory calculation (including the models to incorporate collimator scatter and leaf transmission), (ii) the correct transfer of the leaf sequencing file to the treatment machine, and (iii) the mechanical and dosimetrical performance of the treatment unit. The method was tested for intensity modulated 10 and 25 MV photon beams; both model cases and real clinical cases were studied. Dose profiles measured with the EPID were also compared with ionization chamber measurements. In all cases both predictions and EPID measurements and EPID and ionization chamber measurements agreed within 2% (1 sigma). The study has demonstrated that the proposed method allows fast and accurate pretreatment verification of DMLC.