Taka-aki Hirose

Effect of accounting for interfractional CTV shape variations in PTV margins on prostate cancer radiation treatment plans

PURPOSE The aim of this study was to account for interfractional clinical target volume (CTV) shape variation and apply this to the planning target volume (PTV) margin for prostate cancer radiation treatment plans. METHODS Interfractional CTV shape variations were estimated from weekly cone-beam computed tomography (CBCT) images using statistical point distribution models. The interfractional CTV shape variation was taken into account in the van Herks margin formula. The PTV margins without and with the CTV shape variation, i.e., standard (PTVori) and new (PTVshape) margins, were applied to 10 clinical cases that had weekly CBCT images acquired during their treatment sessions. Each patient was replanned for low-, intermediate-, and high-risk CTVs, using both margins. The dose indices (D98 and V70) of treatment plans with the two margins were compared on weekly pseudo-planning computed tomography (PCT) images, which were defined as PCT images registered using a deformable image registration technique with weekly CBCT images, including contours of the CTV, rectum, and bladder. RESULTS The percentage of treatment fractions of patients who received CTV D98 greater than 95% of a prescribed dose increased from 80.3 (PTVori) to 81.8% (PTVshape) for low-risk CTVs, 78.8 (PTVori) to 87.9% (PTVshape) for intermediate-risk CTVs, and 80.3 (PTVori) to 87.9% (PTVshape) for high-risk CTVs. In most cases, the dose indices of the rectum and bladder were acceptable in clinical practice. CONCLUSION The results of this study suggest that interfractional CTV shape variations should be taken into account when determining PTV margins to increase CTV coverages.

Bayesian delineation framework of clinical target volumes for prostate cancer radiotherapy using an anatomical-features-based machine learning technique

Our aim was to develop a Bayesian delineation framework of clinical target volumes (CTVs) for prostate cancer radiotherapy using an anatomical-features-based machine learning (AF-ML) technique. Probabilistic atlases (PAs) of the pelvic bone and the CTV were generated from 43 training cases. Translation vectors, which could move the CTV PAs to CTV locations, were estimated using the AF-ML after a bone-based registration between the PAs and planning computed tomography (CT) images. An input vector derived from 11 AF points was fed to three AF-ML techniques (artificial neural network: ANN, random forest: RF, support vector machine: SVM). The AF points were selected from edge points and centroids of anatomical structures around prostate. Reference translation vectors between centroids of CTV PAs and CTVs were given to the AF-ML as teaching data. The CTV regions were extracted by thresholding posterior probabilities produced by using the Bayesian inference with the translated CTV PA and likelihoods of planning CT values. The framework was evaluated based on a leave-one-out test with CTV contours determined by radiation oncologists. Average location errors of CTV PAs along the anterior-posterior and superior-inferior directions without AF-ML were 5.7±4.6 mm and 5.5±4.3 mm, respectively, whereas the errors along the two directions with ANN, which showed the best performance, were 2.4±1.7 mm and 2.2±2.2 mm, respectively. The average Dice’s similarity coefficient between reference and estimated CTVs for 44 test cases were 0.81±0.062 with ANN. The framework using AF-ML could accurately estimate CTVs of prostate cancer radiotherapy.

Possibility of chest wall dose reduction using volumetric-modulated arc therapy (VMAT) in radiation-induced rib fracture cases: comparison with stereotactic body radiation therapy (SBRT)

Abstract The present study compares dosimetric parameters between volumetric-modulated arc therapy (VMAT) and 3D conformal radiation therapy (3D-CRT) in lung tumors adjacent to the chest wall treated with stereotactic body radiation therapy (SBRT). The study focused on the radiation dose to the chest wall of 16 patients who had developed radiation-induced rib fractures (RIRF) after SBRT using 3D-CRT. The targets in all patients were partially overlapping with the fractured ribs, and the median overlapping rib–PTV distance was 0.4 cm. Stereotactic body radiation therapy was re-planned for all patients. The prescribed dose was 48 Gy in four fractions to cover at least 95% of the planning target volume (PTV). Evaluated dosimetric factors included D98% and the conformation number (CN) of the PTV, the D2cm3, V40 and V30 of the fractured ribs, the V30 of the chest wall, and the Dmean, V20 and V5 of the lung. A comparison of 3D-CRT with the VMAT plan for PTV revealed that CN was significantly improved in the VMAT plan, whereas D98% did not significantly differ between the two plans. Regarding organs at risk (OARs), the D2cm3, V40 and V30 of fractured ribs, the V30 of the chest wall, and the Dmean, V20 and V5 of the lung, were significantly decreased in the VMAT plan. We concluded that the dose to OARs such as ribs and chest wall could be reduced with improved target conformity using VMAT instead of 3D-CRT for SBRT to treat peripheral lung tumors.

Reconstruction of four-dimensional computed tomography images during treatment time using electronic portal imaging device images based on a dynamic 2D/3D registration

The goal of our study was to develop a computational framework for reconstruction of four-dimensional computed tomography (4D-CT) images during treatment time using electronic portal imaging device (EPID) images based on a dynamic 2D/3D registration. The 4D-CT images during treatment time (“treatment” 4D-CT images) were reconstructed by performing an affine transformation-based dynamic 2D/3D registration between dynamic clinical portal dose images (PDIs) derived from the EPID images with planning CT images through planning PDIs for all frames. Elements of the affine transformation matrices (transformation parameters) were optimized using a Levenberg-Marquardt (LM) algorithm so that the planning PDIs could be similar to the dynamic clinical PDIs for all frames. Initial transformation parameters in each frame should be determined for finding optimum transformation parameters in the LM algorithm. In this study, the optimum transformation parameters in a frame employed as the initial transformation parameters for optimizing the transformation parameter in the consecutive frame. Gamma pass rates (3 mm/3%) were calculated for evaluating a similarity of the dose distributions between the dynamic clinical PDIs and “treatment” PDIs, which were calculated from “treatment” 4D-CT images, for all frames. The framework was applied to eight lung cancer patients who were treated with stereotactic body radiation therapy (SBRT). A mean of the average gamma pass rates between the dynamic clinical PDIs and the “treatment” PDIs for all frames was 98.3±1.2% for eight cases. In conclusion, the proposed framework makes it possible to dynamically monitor patients’ movement during treatment time.

SU-D-BRA-04: Computerized Framework for Marker-Less Localization of Anatomical Feature Points in Range Images Based On Differential Geometry Features for Image-Guided Radiation Therapy

PURPOSE To propose a computerized framework for localization of anatomical feature points on the patient surface in infrared-ray based range images by using differential geometry (curvature) features. METHODS The general concept was to reconstruct the patient surface by using a mathematical modeling technique for the computation of differential geometry features that characterize the local shapes of the patient surfaces. A region of interest (ROI) was firstly extracted based on a template matching technique applied on amplitude (grayscale) images. The extracted ROI was preprocessed for reducing temporal and spatial noises by using Kalman and bilateral filters, respectively. Next, a smooth patient surface was reconstructed by using a non-uniform rational basis spline (NURBS) model. Finally, differential geometry features, i.e. the shape index and curvedness features were computed for localizing the anatomical feature points. The proposed framework was trained for optimizing shape index and curvedness thresholds and tested on range images of an anthropomorphic head phantom. The range images were acquired by an infrared ray-based time-of-flight (TOF) camera. The localization accuracy was evaluated by measuring the mean of minimum Euclidean distances (MMED) between reference (ground truth) points and the feature points localized by the proposed framework. The evaluation was performed for points localized on convex regions (e.g. apex of nose) and concave regions (e.g. nasofacial sulcus). RESULTS The proposed framework has localized anatomical feature points on convex and concave anatomical landmarks with MMEDs of 1.91±0.50 mm and 3.70±0.92 mm, respectively. A statistically significant difference was obtained between the feature points on the convex and concave regions (P<0.001). CONCLUSION Our study has shown the feasibility of differential geometry features for localization of anatomical feature points on the patient surface in range images. The proposed framework might be useful for tasks involving feature-based image registration in range-image guided radiation therapy.

SU-F-T-394: Impact of PTV Margins With Taking Into Account Shape Variation On IMRT Plans For Prostate Cancer

PURPOSE The purpose of this study was to investigate the impact of planning target volume (PTV) margins with taking into consideration clinical target volume (CTV) shape variations on treatment plans of intensity modulated radiation therapy (IMRT) for prostate cancer. METHODS The systematic errors and the random errors for patient setup errors in right-left (RL), anterior-posterior (AP), and superior-inferior (SI) directions were obtained from data of 20 patients, and those for CTV shape variations were calculated from 10 patients, who were weekly scanned using cone beam computed tomography (CBCT). The setup error was defined as the difference in prostate centers between planning CT and CBCT images after bone-based registrations. CTV shape variations of high, intermediate and low risk CTVs were calculated for each patient from variances of interfractional shape variations on each vertex of three-dimensional CTV point distributions, which were manually obtained from CTV contours on the CBCT images. PTV margins were calculated using the setup errors with and without CTV shape variations for each risk CTV. Six treatment plans were retrospectively made by using the PTV margins with and without CTV shape variations for the three risk CTVs of 5 test patients. Furthermore, the treatment plans were applied to CBCT images for investigating the impact of shape variations on PTV margins. RESULTS The percentages of population to cover with the PTV, which satisfies the CTV D98 of 95%, with and without the shape variations were 89.7% and 74.4% for high risk, 89.7% and 76.9% for intermediate risk, 84.6% and 76.9% for low risk, respectively. CONCLUSION PTV margins taking into account CTV shape variation provide significant improvement of applicable percentage of population (P < 0.05). This study suggested that CTV shape variation should be taken consideration into determination of the PTV margins.

WE-AB-207B-03: A Computational Methodology for Determination of CTV-To-PTV Margins with Inter Fractional Shape Variations Based On a Statistical Point Distribution Model for Prostate Cancer Radiation Therapy

PURPOSE Our assumption was that interfractional shape variations of target volumes could not be negligible for determination of clinical target volume (CTV)-to-planning target volume (PTV) margins. The aim of this study was to investigate this assumption as a simulation study by developing a computational framework of CTV-to-PTV margins with taking the interfractional shape variations into account based on point distribution model (PDM) METHODS: The systematic and random errors for interfractional shape variations and translations of target volumes were evaluated for four types of CTV regions (only a prostate, a prostate plus proximal 1-cm seminal vesicles, a prostate plus proximal 2-cm seminal vesicles, and a prostate plus whole seminal vesicles). The CTV regions were delineated depending on prostate cancer risk groups on planning computed tomography (CT) and cone beam CT (CBCT) images of 73 fractions of 10 patients. The random and systematic errors for shape variations of CTV regions were derived from PDMs of CTV surfaces for all fractions of each patient. Systematic errors of shape variations of CTV regions were derived by comparing PDMs between planning CTV surfaces and average CTV surfaces. Finally, anisotropic CTV-to-PTV margins with shape variations in 6 directions (anterior, posterior, superior, inferior, right, and left) were computed by using a van Herk margin formula. RESULTS Differences between CTV-to-PTV margins with and without shape variations ranged from 0.7 to 1.7 mm in anterior direction, 1.0 to 2.8 mm in posterior direction, 0.8 to 2.8 mm in superior direction, 0.6 to 1.6 mm in inferior direction, 1.4 to 4.4 mm in right direction, and 1.3 to 5.2 mm in left direction. CONCLUSION More than 1.0 mm additional margins were needed at least in 3 directions to guarantee CTV coverage due to shape variations. Therefore, shape variations should be taken into account for the determination of CTV-to-PTV margins.

SU-F-J-34: Automatic Target-Based Patient Positioning Framework for Image-Guided Radiotherapy in Prostate Cancer Treatment

PURPOSE Current image-guided radiotherapy (IGRT) procedure is bonebased patient positioning, followed by subjective manual correction using cone beam computed tomography (CBCT). This procedure might cause the misalignment of the patient positioning. Automatic target-based patient positioning systems achieve the better reproducibility of patient setup. Our aim of this study was to develop an automatic target-based patient positioning framework for IGRT with CBCT images in prostate cancer treatment. METHODS Seventy-three CBCT images of 10 patients and 24 planning CT images with digital imaging and communications in medicine for radiotherapy (DICOM-RT) structures were used for this study. Our proposed framework started from the generation of probabilistic atlases of bone and prostate from 24 planning CT images and prostate contours, which were made in the treatment planning. Next, the gray-scale histograms of CBCT values within CTV regions in the planning CT images were obtained as the occurrence probability of the CBCT values. Then, CBCT images were registered to the atlases using a rigid registration with mutual information. Finally, prostate regions were estimated by applying the Bayesian inference to CBCT images with the probabilistic atlases and CBCT value occurrence probability. The proposed framework was evaluated by calculating the Euclidean distance of errors between two centroids of prostate regions determined by our method and ground truths of manual delineations by a radiation oncologist and a medical physicist on CBCT images for 10 patients. RESULTS The average Euclidean distance between the centroids of extracted prostate regions determined by our proposed method and ground truths was 4.4 mm. The average errors for each direction were 1.8 mm in anteroposterior direction, 0.6 mm in lateral direction and 2.1 mm in craniocaudal direction. CONCLUSION Our proposed framework based on probabilistic atlases and Bayesian inference might be feasible to automatically determine prostate regions on CBCT images.

SU-E-J-177: A Computational Approach for Determination of Anisotropic PTV Margins Based On Statistical Shape Analysis for Prostate Cancer Radiotherapy

Purpose: Our aim of this study was to propose a computational approach for determination of anisotropic planning target volume (PTV) margins based on statistical shape analysis with taking into account time variations of clinical target volume (CTV) shapes for the prostate cancer radiation treatment planning (RTP). Methods: Systematic and random setup errors were measured using orthogonal projection and cone beam computed tomography (CBCT) images for data of 20 patients, who underwent the intensity modulated radiation therapy for prostate cancer. The low-risk, intermediate-risk, and high-risk CTVs were defined as only a prostate, a prostate plus proximal 1-cm seminal vesicles, and a prostate plus proximal 2-cm seminal vesicles, respectively. All CTV regions were registered with a reference CTV region with a median volume to remove the effect of the setup errors, and converted to a point distribution models. The systematic and random errors for translations of CTV regions were automatically evaluated by analyzing the movements of centroids of CTV regions. The random and systematic errors for shape variations of CTV regions were obtained from covariance matrices based on point distributions for the CTV contours on CBCT images of 72 fractions of 10 patients. Anisotropic PTV margins for 6 directions (right, left, anterior, posterior, superior and inferior) were derived by using Yoda’s PTV margin model. Results: PTV margins with and without shape variations were 5.75 to 8.03 mm and 5.23 to 7.67 mm for low-risk group, 5.87 to 8.33 mm and 5.23 to 7.67 mm for intermediate-risk group, and 5.88 to 8.25 mm and 5.29 to 7.82 mm for highrisk group, respectively. Conclusion: The proposed computational approach could be feasible for determination of the anisotropic PTV margins with taking into account CTV shape variations for the RTP.