Setup deviations in wedged pair irradiation of parotid gland and tonsillar tumors, measured with an electronic portal imaging device.

The first aim of this study was to quantify estimated translational setup deviations of patients treated with a wedged pair of oblique beams for parotid gland and tonsillar tumors, using portal imaging. The second aim was to design an off-line setup verification procedure, to improve the setup accuracy, if necessary. Thirty-one patients were treated with two conformal fields (anterior-oblique and posterior-oblique). The patients were immobilized with a head cast. For the last 10 patients, the rigidity of the cast was improved while, in addition, wax molds with metal markers were placed into the outer ear for image correlation. Portal images were acquired about weekly. Setup deviations were analyzed, using anatomical structures and, when available, metal markers for image matching. The consistency of the deviations was determined by the correlation between deviations in the cranio-caudal direction, as measured from both beams. When the deviations were consistent, the translational setup deviation during a treatment session could be described by a three-dimensional (3D) vector. A setup verification procedure was designed using a computer simulation. The statistics of the 3D setup deviations were used as input. The output consisted of the resulting setup accuracy and workload (i.e., the number of setup corrections and portal images). Using the anatomical structures for image correlation, the deviations in the cranio-caudal direction were not correlated, either for the old or the improved cast. However, by using the metal markers, the deviations were correlated and a 3D analysis could be performed. The standard deviations, averaged over the three directions, were equal to 1.8 and 1.4 mm for the distribution of systematic and random deviations, respectively. Application of a setup verification procedure, with 0.7 corrections on the average per patient, could potentially reduce the percentage of 3D systematic deviations larger than 4 mm from 30 to 2%. It can be concluded that it was not possible to obtain consistent translational setup deviations, due to rotations. To quantify 3D translational setup deviations, it was necessary to use additional metal markers, which were visible in the portal images of both beams. A further improvement of the setup accuracy is possible by using an off-line setup verification procedure.

Optimization of automatic portal image analysis.

The purpose of this study is to quantify and optimize the performance of an automatic portal image analysis procedure under clinical conditions and to compare the performance with that of human operators. A new method, based on analysis of variance, is introduced to quantify the clinical performance of portal image analysis tools in terms of systematic and random variations. The automatic portal image analysis procedure is based on chamfer matching. Two image enhancement techniques have been investigated in the automatic procedure: morphological top-hat (MTH) transformation and multiscale medial axis (MMA) transformation. The performance of these enhancements was quantified and optimized as a function of filter size using images obtained from clinical treatment. All images used for this study were obtained from pelvic treatment fields by means of an electronic portal imaging device. The random variations in the alignment of AP fields are typically 0.5 mm and 0.5 degrees (1 SD) for both the human operators and the optimized automatic analysis procedure. Random variations in the alignment of lateral pelvic fields are typically twice as large for all operators. MMA enhancement yields smaller random variations than MTH enhancement for lateral fields, but the differences are marginal for AP fields. The optimized automatic analysis procedure has a success rate ranging from 99% for AP large fields to 96% for lateral fields and 85% for AP boost fields. The accuracy of the method is comparable with the accuracy of the human operators for most investigated fields. For lateral boost fields and simultaneous boost fields, the random variations of the automatic analysis are typically two times larger than the variations of the human operators. Automatic analysis is 4 to 20 times faster than human operators yielding a large reduction in work load.

In vivo determination of the accuracy of field matching in breast cancer irradiation using an electronic portal imaging device.

The purpose of this study was to investigate the accuracy of field matching in patients treated by irradiation of the breast and adjacent lymph nodes. Field matching is performed by the radiographers during each session on a match line drawn on the patient's skin. Field edge positions were assessed in the cranial match plane of tangential breast fields and supraclavicular-axillary fields using an electronic portal imaging device and match line markers placed on the skin of the patients. The mean gap/overlap of the four fields for individual patients during each treatment session, derived from 374 marker projections, was +0.5 mm indicating that no systematic gap or overlap was observed. The uncertainty in the position of the four fields with respect to the match plane ranges from 3.1 to 5.1 mm (1 SD) for the individual patients. Gaps and overlaps between fields were also related to an absolute match line position, found by comparison of simulator and portal images, showing a small systematic uncertainty of 2.4 mm and a standard deviation of 3.3 mm. It can be concluded that the use of an electronic portal imaging device in combination with match line markers is a good method to quantify the accuracy of field matching in vivo. The results showed good stability and reproducibility in the field matching region for this treatment technique of breast cancer irradiation.

Transfer errors of planning CT to simulator: a possible source of setup inaccuracies?

The purpose of this study was to analyse whether the intended patient setup, based on a CT scan, was different from the setup at the simulator. Furthermore, we investigated how these possible transfer errors between the planned patient setup and the actual simulator setup affected the resulting overall treatment setup accuracy. Two groups, of 15 prostate patients each, were studied. For one group (group II), the simulation time was about twice as large as for the other (group I), since digitally reconstructed radiographs (DRRs) were used to get a good visual agreement between the intended and the simulator setup. For the purpose of this study DRRs were also calculated for the patients in group I, and for both groups DRRs were matched with the simulator images to obtain quantitative data of the transfer errors. The resulting overall treatment setup accuracy was determined by comparing the DRRs with portal images. For group I, the standard deviations (SD) of the differences between the DRRs and the simulator images ('transfer errors') were 1.5 mm and 4.5 mm in the lateral (x) and cranio-caudal (y) direction, respectively. For group II the SDs were smaller: 1.4 mm and 1.5 mm in the x- and y-direction, respectively. For both groups, the magnitude of the overall mean was less than 1.3 mm. For group I, the SDs of the resulting overall setup deviations during treatment were 1.6 mm and 4.1 mm in the x- and y-direction, respectively. For group II, these figures were 2.4 mm and 2.6 mm, respectively. For both groups, the magnitude of the overall mean was less than 1.0 mm.(ABSTRACT TRUNCATED AT 250 WORDS)

A comprehensive system for the analysis of portal images.

In recent years, several techniques for the processing and analysis of portal images have been developed. It is the aim of this study to integrate some of these techniques into one comprehensive system. An advantage of this approach is that clinical experience can be obtained with more than one technique and a comparison of the techniques becomes possible. The portal image analysis procedure is implemented in the following steps: preparation of the reference image, portal image field edge detection, field edge match, anatomy match and the presentation of the results. For most of these steps, several alternative methods (e.g., interactive and automatic) are implemented. In addition, two new visualisation techniques have been incorporated. The first is a method for combining the results of the analysis of multiple fields in two dimensions, e.g., large and boost fields. The second is a method for three-dimensional reconstruction of beam setup data, as derived from portal image analysis, on arbitrary reconstructed slices of a CT scan. With the latter method, the effect of setup errors on complex treatments (e.g., matching fields) can be studied. The new system has been in clinical use in our institution for two years and has been used to analyse about 5000 clinical portal images. The operators could choose freely from several matching methods. For 83% of the images our automatic matching algorithm was used. When required, the result of this method was corrected using the interactive drawing on image match. Significant corrections (more than 1 mm translation or 1 degree rotation) were applied to 27% of the automatically analysed images.(ABSTRACT TRUNCATED AT 250 WORDS)

An algorithm for automatic analysis of portal images: clinical evaluation for prostate treatments.

The aim of this study is to assess the clinical value of an algorithm for automatic analysis of portal images by measuring the method's performance in a clinical study of treatment of prostate cancer. The algorithm is based on chamfer matching and measures displacements of patients relative to prescribed radiation beam positions. In this paper we propose a method to quantify the mean standard deviation (MSD) of the performance of automatic analysis relative to the MSD of the performance of trained radiographers using the clinical data set only, i.e. without using additional phantoms or simulations. The clinical data set in this study consists of 99 regional AP prostate images of 15 different patients. To assess the performance the automatic analysis in relation to that of the human observers, we studied the results of the unsupervised automatic analysis, as well as the results of a less-trained human observer and a well-trained human observer assisted by the automatic analysis (in this combination, automatic analysis is done first and the result is modified by the well-trained observer if the observer does not agree). First, the intra-observer variations of the well-trained observer are measured by repetitive analysis of a small subset of the clinical data. The distribution of differences in analysis between two arbitrary observers is described by the chi 2 distribution, and is tabulated in literature. We define the agreement histogram of an observer O as an estimator for the chi 2 distribution between O and the well-trained human observer, parameterized by the ratio of the intra-observer variations of O and the well-trained observer.(ABSTRACT TRUNCATED AT 250 WORDS)

Maximizing setup accuracy using portal images as applied to a conformal boost technique for prostatic cancer.

A design procedure of a patient setup verification protocol based upon frequent digital acquisition of portal images is demonstrated with an application for conformal prostatic boost fields. The protocol aims at the elimination of large systematic deviations in the patient setup and includes decision rules which indicate when correction of the patient setup is needed. The decision rules were derived from the results of a theoretical and quantitative analysis of patient setup variations measured in three pelvic fields (one anterior-posterior and two lateral fields) of 105 fractions for nine patients. Deviations in the patient positioning, derived from one field, were quantified as two-dimensional (2-D) displacement vectors in the plane perpendicular to the beam axis by alignment of anatomical features in the portal and the simulator image. The magnitude of the overall setup variations along the anterior-posterior, superior-inferior and lateral directions varied between 2.6 and 3 mm (1 S.D.). Inter- and intra-treatment variations could be separated, both having equal magnitudes of 1.7 to 2.2 mm (1 S.D.). In addition, intra-treatment variations appeared to be predictable which was a prerequisite for the development of the decision rules. The 2-D setup deviations, measured in the three fields of one fraction were strongly correlated and a 3-D displacement vector was calculated. Utilization of this 3-D vector in a setup verification protocol may lead to an early detection of systematic setup deviations.