Search

A generic method for three-dimensional (3-D) evaluation of target volume delineation in multiple imaging modalities is presented. The evaluation includes geometrical and statistical methods to estimate observer differences and variability in defining the Gross Tumor Volume (GTV) in relation to the diagnostic CT and MRI modalities. The geometrical method is based on mapping the 3-D shape of the target volume to a scalar representation, thus enabling a one-dimensional statistical analysis. The statistical method distinguishes observer and modality related uncertainties, which are expressed in terms of three error components: random observer deviations, systematic observer differences, and systematic modality differences. Monte Carlo simulations demonstrate that the standard errors of each of the three model parameters are inversely proportional to the square root of the product of the patient group size and the number of observers and proportional to the intraobserver variation. For 18 patients and 3 observers the standard errors of the estimated systematic modality and observer differences are 19% and 14% of the intraobserver standard deviation, respectively. A scalar representation of the shape of the prostate, delineated by 3 observers for 18 patients, was obtained by sampling the distance between the average center of gravity of the prostate in CT and the prostate surface for a large number of directions (2500), using polar coordinates. Observer variability and differences were obtained by applying the statistical method to the samples independently. The intraobserver variation for CT was largest in regions near the seminal vesicles (s.d: 3 mm) and the apex (s.d: 3 mm). The systematic observer variation in CT was largest in a region near the plexus Santorini, at the caudal-anterior side of the prostate (s.d.: 2 mm). The sensitivity for the choice of origin was tested by using the average center of gravity from axial MRI instead of CT. The results were almost identical. The polar map measures distances in the scanning directions. A correction procedure to get the variability in directions perpendicular to the surface of the prostate yielded variations that were a factor of 0.85 smaller for all directions. It is concluded that by separating the shape evaluation in a geometrical and a statistical part, the complexity of the analysis of 3-D shape differences can be significantly reduced. The method was successfully applied to a group of prostate patients, where we demonstrated that delineation variability is nonhomogeneous, with the largest variations occurring near the seminal vesicles and the apex.

The purpose of this study is to develop a method for registration of CT and MR scans of the pelvis with minimal user interaction and to obtain a means for objective quantification of the registration accuracy of clinical data without markers. CT scans were registered with proton density MR scans using chamfer matching on automatically segmented bone. A fixed threshold was used to segment CT, while morphological filters were used to segment MR. The method was tested with transverse and coronal MR scans of 18 patients and sagittal MR scans of 8 patients. The registration accuracy was estimated by comparing (triangulating) registrations of a single CT scan with MR in different orientations in a "full circle." For example, CT is first matched on transverse MR, next transverse MR is matched independently on coronal MR, and finally coronal MR is matched independently on CT. The product of the three transformations is the identity if all matching steps are perfect. Deviations from identity occur both due to random errors and due to some types of systematic errors. MR was registered on MR (to close the "circle") by minimization of rms voxel value differences. CT-MR registration takes about 1 min, including user interaction. The random error for CT-MR registration with transverse or coronal MR was 0.5 mm in translation and 0.4 degree in rotation (standard deviation) for each axis. A systematic registration error of about 1 mm was demonstrated along the MR frequency encoding direction, which is attributed to the chemical shift. In conclusion, the presented algorithm efficiently and accurately registers pelvic CT and MR scans on bone. The "full circle" method provides an estimate of the registration accuracy on clinical data.

In the treatment planning of conformal radiotherapy, field shapes are often designed in such a way that a high-value isodose surface fully encompasses the target volume. Therefore, knowledge about the accuracy with which the treatment planning system calculates the position of that isodose surface is essential to prevent field shapes which are either too large or too small. To determine this accuracy for a conformal multi-field technique, the dose in the high-dose region must be measured with a high spatial resolution. A method is presented to reconstruct and evaluate the experimental high-dose region from a set of water phantom scans. This method, which assesses combined dose profiles for multi-field irradiation techniques, can be used for the commissioning and/or quality assurance of a 3-D treatment planning system. For a specific conformal technique, the measured and calculated 95% isodose positions along lines in several directions have been compared. It is shown that different dose values of single beam profiles determine the resulting 95% isodose position, which is important to recognize for quality assurance of treatment planning calculations. It is further found that the uncertainty in the calculated 95% isodose surface can be described by a standard deviation in dose value, which relates to a positional uncertainty through the local dose gradient. Thus the confidence region of the calculated 95% isodose can be indicated in the treatment plan by plotting isodoses at the 95% level plus and minus its standard deviation. Such a procedure is recommended instead of plotting the 95% isodose with a constant width. In addition, restrictions for the cumulative dose-volume histogram of acceptable treatment plans can be formulated, based on the sensitivity of the actual target coverage on the uncertainty with which the prescribed isodose surface is calculated.

In this study we investigate a method for positioning the margin required around the clinical target volume (CTV) to account for the random geometrical treatment uncertainties during conformal radiotherapy. These uncertainties are introduced by patient setup errors and CTV motion within the patient. Three-dimensional dose distributions are calculated for two four-field box techniques and a three-field technique, using rectangular fields. In addition, dose calculations are performed for four prostate cases, treated with a three-field conformal technique. The effects of random rotational and translational deviations on the delivered dose are described as a convolution of the "static" dose with the distribution of the deviations. For the rectangular field techniques, these convolutions are performed with a range of standard deviations (SDs) of the distribution of random translations (0-7 mm in the three directions) and rotations (0 degree-5 degrees around the main axes). Two centers of rotation are considered: the isocenter and a position that is 3.5 cm shifted with respect to the isocenter. For the prostate cases, the random deviations are estimated by combining the results from organ motion and setup accuracy studies. The required margin is defined as the change in the position of the static 95% isodose surface by the convolution and it is approximated by a morphological erosion operator, applied to the static 95% isodose surface. When the center of rotation coincides with the isocenter the change in the position of the static 95% isodose surface can accurately be described by an erosion operator. For the rectangular field techniques, the margin is equal to about 0.7 SD of the distribution of translations, independent of the distribution of rotations. When the center of rotation does not coincide with the isocenter and rotations are considerable, the margin is strongly place dependent, and the accuracy of the approximation by an erosion operator is much lower. In conclusion, margins for random uncertainties can be approximated by a dilation operator (inverse of an erosion operator) when the center of rotational deviations coincides with the isocenter. The size of the margin is about 0.7 SD of the distribution of translations. When rotational deviations are present and the center of rotation does not coincide with the isocenter, the margin can become strongly place dependent and the convolution computation should be incorporated in the planning system.