Y. Chi

Adaptive Replanning Strategies Accounting for Shrinkage in Head and Neck IMRT

PURPOSE Significant anatomic and volumetric changes occur in head and neck cancer patients during fractionated radiotherapy, and the actual dose can be considerably different from the original plan. The purposes of this study were (1) to evaluate the differences between planned and delivered dose, (2) to investigate margins required for anatomic changes, and (3) to find optimal replanning strategies. METHODS AND MATERIALS Eleven patients, each with one planning and six weekly helical CTs, were included. Intensity-modulated radiotherapy plans were generated using the simultaneous integrated boost technique. Weekly CTs were rigidly registered to planning CT before deformable registration was performed. The following replanning strategies were investigated with different margins (0, 3, 5 mm): midcourse (one replan), every other week (two replans), and every week (six replans). Doses were accumulated on the planning CT for comparison of various dose indices for target and critical structures. RESULTS The cumulative doses to targets were preserved even at the 0-mm margin. Doses to cord, brainstem, and mandible were unchanged. Significant increases in parotid doses were observed. Margin reduction from 5 to 0 mm led to a 22% improvement in parotid mean dose. Parotid sparing could be preserved with replanning. More frequent replanning led to better preservation; replanning more than once a week is unnecessary. CONCLUSION Shrinkage does not result in significant dosimetric difference in targets and critical structures, except for the parotid gland, for which the mean dose increases by approximately 10%. The benefit of replanning is improved sparing of the parotid. The combination of replanning and reduced margins can provide up to a 30% difference in parotid dose.

A material sensitivity study on the accuracy of deformable organ registration using linear biomechanical models

Model-based deformable organ registration techniques using the finite element method (FEM) have recently been investigated intensively and applied to image-guided adaptive radiotherapy (IGART). These techniques assume that human organs are linearly elastic material, and their mechanical properties are predetermined. Unfortunately, the accurate measurement of the tissue material properties is challenging and the properties usually vary between patients. A common issue is therefore the achievable accuracy of the calculation due to the limited access to tissue elastic material constants. In this study, we performed a systematic investigation on this subject based on tissue biomechanics and computer simulations to establish the relationships between achievable registration accuracy and tissue mechanical and organ geometrical properties. Primarily we focused on image registration for three organs: rectal wall, bladder wall, and prostate. The tissue anisotropy due to orientation preference in tissue fiber alignment is captured by using an orthotropic or a transversely isotropic elastic model. First we developed biomechanical models for the rectal wall, bladder wall, and prostate using simplified geometries and investigated the effect of varying material parameters on the resulting organ deformation. Then computer models based on patient image data were constructed, and image registrations were performed. The sensitivity of registration errors was studied by perturbating the tissue material properties from their mean values while fixing the boundary conditions. The simulation results demonstrated that registration error for a subvolume increases as its distance from the boundary increases. Also, a variable associated with material stability was found to be a dominant factor in registration accuracy in the context of material uncertainty. For hollow thin organs such as rectal walls and bladder walls, the registration errors are limited. Given 30% in material uncertainty, the registration error is limited to within 1.3 mm. For a solid organ such as the prostate, the registration errors are much larger. Given 30% in material uncertainty, the registration error can reach 4.5 mm. However, the registration error distribution for prostates shows that most of the subvolumes have a much smaller registration error. A deformable organ registration technique that uses FEM is a good candidate in IGART if the mean material parameters are available.

Comparison of Planned Versus Actual Dose Delivered for External Beam Accelerated Partial Breast Irradiation Using Cone-Beam CT and Deformable Registration

PURPOSE To assess the adequacy of dose delivery to the clinical target volume (CTV) using external beam (EB) accelerated partial breast irradiation (APBI). METHODS AND MATERIALS Sixteen patients treated with EB APBI underwent cone beam CT (CBCT) before each fraction and daily helical CT (HCT) scans to determine setup errors and calculate the dose per fraction. For 12 patients, an in-house image-intensity-based deformable registration program was used to register the HCTs to the planning CT and generate the cumulative dose. Treatment was 38.5 Gy in 10 fractions. EB APBI constraints from the National Surgical Adjuvant Breast and Bowel Project B39/Radiation Therapy Oncology Group 0413 Phase III protocol were used. RESULTS The mean setup error per CBCT registration was 9 ± 5 mm. Dose-volume histogram analysis showed only one patient (8%) with a decrease in the CTV V90 (8% underdosage). All other patients demonstrated adequate target coverage. PTV_EVAL V90 was on average 3% (range, 0%-16%) less than planned. For the ipsilateral breast, four patients had an increase in V50 (≤ 1% increase) and three patients had an increase in V100 (≤ 9% increase). Only one patient showed an increase >5%. Four patients had an increase in ipsilateral lung V30 (maximum 3%), and one had an increase in heart V5 (1%). Four patients had an increase in MaxDose (maximum 89 cGy). CONCLUSIONS The current CTV-to-PTV margin of 10 mm appears sufficient for ∼92% of patients treated with EB APBI. Although expansion of the population PTV margin to 14 mm would provide ∼97% confidence level for CTV coverage, online image guidance should be considered.

SU‐FF‐T‐163: Dose Calculation On Cone Beam CT (CBCT)

Purpose:CBCT provides useful information for image guidance, however, uncertainty in CT number prevents it from being used for accuracy dose calculation. The purpose of this study is to quantify the accuracy of dose calculation using different CT to density mapping methods and investigate achievable accuracy when applying these methods for dose calculation based on CBCT.Method and Materials: The heterogeneous dose calculated on helical CT (HCT) is used as reference, and the resulting dose from other three methods are compared to reference. In homogeneous dose calculation (HO) method, the density value inside skin contour is overwritten as 1g/cm3. In air‐bone‐soft tissue (ABS) method, the source CTimage were replaced by three regions, air, bone, and soft tissue with density overwritten as 0.1 g/cm3, 1.4 g/cm3, and 1 g/cm3 respectively. In stepwise CT density (SWD) method, a CT‐density table representing a stepwise function is applied to CT and used in dose calculation. The three methods are applied to both HCT and CBCT on Pinnacle3 planning system. The resulting dose distributions are compared. Results: Head and neck patients are selected for study. For dose calculation on HCT, the maximum discrepancies in mean dose for HO, ABS, and SWD are within 4%, 2%, and 0.3% for critical organs respectively, and within 1%, 0.5%, and 0.5% for targets. The accuracy results for ABS can be extended to CBCT. For dose calculation on CBCT, dose distribution from ABS is used as reference. The maximum discrepancy between HO and ABS is 2% for critical organs and 0.5% for targets; the maximum discrepancy between SWD and ABS is %1 for critical organs, and 0.5% for targets. Conclusions: Both ABS and SWD can achieve 2% accuracy in dose calculation on CBCT. No contour information is required for SWD, however, normalization of CT number may be necessary.

SU-FF-J-34: Can 4D Dose Be Constructed Without Using Deformable Organ Registration?

Purpose:Image based deformable organ registration has been the primary means for 4D dose construction of moving organ. However, patient free breathing cone beam CT (FB‐CBCT) is more routinely obtained, instead of 4D CBCT. In this study, we examine if FB‐CBCT imaging can be directly applied for 4D dose construction. Method and Materials:Organs with the mean shape obtained from 4D reference CTs were mapped on a FB‐CBCT image by applying rigid‐body matching. The 4D dose in the organ was then constructed on the mean CTimage using the 3‐dimenional‐motion pdf detected by the CB projection images. Three lungcancer patients with a planning and multiple daily 4D CTs were used to evaluate this method. The daily FB‐CBCT was constructed using the daily 4D CT. For each patient, dose distributions obtained from a 3D inverse planning and a 4D inverse planning were used in the evaluation. Dose distribution constructed using the mean organ and motion pdf was compared to the 4D dose calculated using the deformable organ registration on the daily 4D CTs. The minimal dose (D99) and EUD for the target, and the maximum dose (D1) for critical structures were used in the comparison. Results: Dose discrepancy (the mean ± SD) for D99 and EUD of the targets was −0.6%±2.6% and 0.0%±1.0%. Dose discrepancy of D1 was 1.5%±1.5% for heart, −1.6%±3.3% for Aorta and −0.6%±3.7% for Esophagus. Dose discrepancy in the lung was not evaluated in the current study due to the need of using variable pdfs at different location of the lung.Conclusion: 4D dose in organs of interest could be directly constructed using daily FB‐CBCT and measured pdf. The construction accuracy seems acceptable for the organs of interest inside of the lung. Support in part by Elekta Research Grant

SU‐GG‐J‐42: Automatic Contour Delineation On Cone Beam CT (CBCT) and Verification

Purpose: Patient anatomy manifested on cone beam CT(CBCT)image is useful for treatment localization, however, automatic organ segmentation on CBCT is challenging and its accuracy needs to be established. The purpose of this study is to quantify the accuracy of automatic contour delineation on CBCT using weekly helical CT (HCT) and daily CBCTimages.Method and Materials:Images from 5 head and neck IMRT patients were used in this study. Each patient had 5 to 6 weekly HCTs and daily CBCTs. Those images are registered to planning HCT using a freeform deformable image registration algorithm. Contours were automatically generated for both HCTs and CBCTs using planning contours and registered displacement. The resulting contours on weekly HCT were used as reference, and compared to contours on CBCT with respect to ROI volume, ROI center coordinate, and ROI surface discrepancy. The ROI surface discrepancy was determined using distance transform of ROI masks. Results: GTV, left parotid, right parotid and mandible, are selected for comparison. GTV volume discrepancy in percentage is −2.8±6.1%, ranging from −9.8% to 11.5 %. GTV volume discrepancy in magnitude is −2.3±3.6cc, ranging from −7.8 to 3.8cc. Volume discrepancies for left parotid, right parotid, and mandible are −6.7±4.7%, −3.6±4.1%, −4.3±5.0% in percentage respectively, and −1.3±1.1cc, −3.2±3.6cc, −0.8±1.0cc in magnitude. The mean difference in ROI centers for all organs is less than 2mm with a maximum of 4mm. Surface discrepancy of GTV, left parotid, right parotid, and mandible are −0.33±1.8mm, −0.28±1.1mm, −0.32±0.9mm, and −0.28±0.9mm respectively. Conclusion: Most of contours from CBCT has slightly smaller volume but all show excellent match to those from HCT, with most surface discrepancy within image voxel size. Further investigations on quantifying dosimetric effect of the geometry uncertainty are underway. Conflict of Interest: Support in part by NCI Grant — CA091020.

SU‐GG‐J‐130: Monte Carlo Simulation of Elekta XVI Cone Beam System and Dose Distribution in CTDI Phantom

Purpose: (1) To validate a Monte Carlo model for kV cone beam CT simulations. (2) To determine whether a bow‐shaped radial dose distribution with equal weighting can be applicable for cone beam CT.Method and Materials: A BEAMnrcMP Monte Carlo package has been employed and modified to simulate the Elekta XVI system operated at 120 kVp with collimator M20 and bowtie filter F1. Percentage Depth Dose (PDD) has been measured with PTW Farmer and Roos chambers in a CNMC water tank as well as a GAMMEX Solid Water phantom at 100cm SSD with 10cm backscatter. Dose profile at depth of 1cm has been measured with a Wellhofer IC‐10 chamber in Solid Water. Sample dose calculation is performed on a standard CDTI phantom aligned at Linac isocenter. Results: The simulated PDD curve agrees very well with both Farmer and Roos chamber measurements in water, but differ slightly with the Farmer chamber measurement in solid water at large depths and build‐up region. This is due to the higher attenuation of solid water in kV range and the loss of scatter equilibrium at phantom surface. Excellent agreement has been shown between the measured dose profile and the Monte Carlo simulations along both in‐plane and cross‐plane directions. The CTDI calculations have indicated that equal weighting on both the center and the periphery can result in errors of −1.1% while unequal standard weighting (1/3 for center and 2/3 for periphery) can result in errors of up to 8.1%. Conclusion: Our Monte Carlo model has been validated by various measurements to simulate the kV cone beams from the Elekta XVI system. A bow‐shape radial profile in CDTI phantom results in equal weightings for volumetric average dose calculations using only center and peripheral CTDI measurements. Investigation on patient dose deposition due to kV CBCT is in progress.

TH‐D‐M100F‐05: Online Region‐Of‐Interest Delineation of Daily Head and Neck Images

Purpose: Online imaging modalities, such as cone beam computed tomography(CBCT) or CT on‐rail provide online volumetric images. A fast, automatic and robust region‐of‐interest (ROI) delineation method is highly desired in image guided radiation therapy(IGRT). We have developed such a method and tested it via segmentation of head and neck (HN) fan beam CT and CBCTimages.Material and Methods: ROIs on planning CTimages were manually delineated using commercial treatment planning system. A variational‐based deformable image registration algorithm was implemented to register planning CTimages to daily CTimages. ROIs on planning CTimages were automatically mapped to daily images using voxel matching information between planning and daily image datasets. The results were quantitatively and qualitatively validated by comparing to manual delineation. In order to accelerate computing speed, we paralleled the algorithm using message passing interface (MPI) on a Beowulf cluster with 16 processing elements (PE). Speed improvement was benchmarked. Results: The discrepancies between automatically and manually delineated ROIs on fan beam images were mostly within 2mm. Automatic segmentation of CBCTimages was acceptable by visual inspection. Benchmark results showed that paralleling efficiencies were above 95% and speedup factors were approximately equal to the number of PE used. With 16 PEs online delineation of HN images took about 1 minute. Conclusion: The online ROI delineation method we have developed is robust, fast and is suitable for HN online adaptive radiation treatments. This research was partially supported by the Department of Defense Prostate Cancer Research Program under award number W81XWH‐07‐0083.

SU‐FF‐J‐73: Non‐Rigid Setup Errors in HN‐IMRT Patients and Their Dosimetric Effect

Purpose: Dose distribution in head and neck IMRT plan is highly conformal and therefore sensitive to setup error. Rigid setup error measurements and its dosimetric effects have been reported previously. In this study, we investigate the non‐rigid component caused by the flexible bony structures in the neck region. The purposes are to define and measure the non‐rigid setup errors throughout the treatment course and to study its dosimetric effect and margin implications. Method and Materials: Daily cone beam CT(CBCT) was acquired for patients receiving HN‐IMRT treatment. For each CBCT three regions (head, neck and shoulder) were rigidly registered to their corresponding part in planning CT individually. We define the non‐rigid setup error as the difference between the maximum and minimum of the translation/rotation variables among three registrations. A zero value indicates a non‐existent non‐rigid setup error. To model the dosimetric effect, we mathematically transformed helical planning CT by keeping shoulder still, rigidly rotating head by ±10 degrees in three directions and deforming neck regions to match head and shoulder. The original plan was applied to these deforming CTs. Results: The non‐rigid setup error is larger in last week than first week, also larger in second half than first half of treatment course, probably due to the weight loss and the mask getting loose. The rotations in L‐R axis (4°) and translations in S‐I direction (5 mm) are larger than others. Margins of 5 mm used in treatment planning are adequate for most organs to account for the non‐rigid setup errors. Conclusion: We have measured non‐rigid setup errors in HN‐IMRT patients using daily CBCTs. We found that rotations in L‐R axis and translations in S‐I direction are dominant and they increase during the treatment course. We also developed a technique to study its dosimetric effect by transforming/deforming planning CTimage.