MU variability in CBCT-guided online adaptive radiation therapy.

PURPOSE: CBCT-guided online-adaptive radiotherapy (oART) systems have been made possible by using artificial intelligence and automation to substantially reduce treatment planning time during on-couch adaptive sessions. Evaluating plans generated during an adaptive session presents significant challenges to the clinical team as the planning process gets compressed into a shorter window than offline planning. We identified MU variations up to 30% difference between the adaptive plan and the reference plan in several oART sessions that caused the clinical team to question the accuracy of the oART dose calculation. We investigated the cause of MU variation and the overall accuracy of the dose delivered when MU variations appear unnecessarily large. METHODS: Dosimetric and adaptive plan data from 604 adaptive sessions of 19 patients undergoing CBCT-guided oART were collected. The analysis included total MU per fraction, planning target volume (PTV) and organs at risk (OAR) volumes, changes in PTV-OAR overlap, and DVH curves. Sessions with MU greater than two standard deviations from the mean were reoptimized offline, verified by an independent calculation system, and measured using a detector array. RESULTS: MU variations relative to the reference plan were normally distributed with a mean of -1.0% and a standard deviation of 11.0%. No significant correlation was found between MU variation and anatomic changes. Offline reoptimization did not reliably reproduce either reference or on-couch total MUs, suggesting that stochastic effects within the oART optimizer are likely causing the variations. Independent dose calculation and detector array measurements resulted in acceptable agreement with the planned dose. CONCLUSIONS: MU variations observed between oART plans were not caused by any errors within the oART workflow. Providers should refrain from using MU variability as a way to express their confidence in the treatment planning accuracy. Clinical decisions during on-couch adaptive sessions should rely on validated secondary dose calculations to ensure optimal plan selection.

How Does the Number of Brain Metastases Correlate With Normal Brain Exposure in Single-Isocenter Multitarget Multifraction Stereotactic Radiosurgery.

Purpose: To investigate the relationship between normal brain exposure in LINAC-based single-isocenter multitarget multifraction stereotactic radiosurgery or stereotactic radiation therapy (SRT) and the number or volume of treated brain metastases, especially for high numbers of metastases. Methods and Materials: A cohort of 44 SRT patients with 709 brain metastases was studied. Renormalizing to a uniform prescription of 27 Gy in 3 fractions, normal brain dose volume indices, including V23 Gy (volume receiving >23 Gy), V18 Gy (volume receiving >18 Gy), and mean dose, were evaluated on these plans against the number and the total volume of targets for each plan. To compare with exposures from whole-brain radiation therapy (WBRT), the SRT dose distributions were converted to equivalent dose in 3 Gy fractions (EQD3) using an alpha-beta ratio of 2 Gy. Results: With increasing number of targets and increasing total target volume, normal brain exposures to dose ≥18 Gy increases, and so does the mean normal brain dose. The factors of the number of targets and the total target volume are both significant, although the number of targets has a larger effect on the mean normal brain dose and the total target volume has a larger effect on V23 Gy and V18 Gy. The EQD3 mean normal brain dose with SRT planning is lower than conventional WBRT. On the other hand, SRT results in higher hot spot (ie, maximum dose outside of tumor) EQD3 dose than WBRT. Conclusions: Based on clinical SRT plans, our study provides information on correlations between normal brain exposure and the number and total volume of targets. As SRT becomes more greatly used for patients with increasingly extensive brain metastases, more clinical data on outcomes and toxicities is necessary to better define the normal brain dose constraints for high-exposure cases and to optimize the SRT management for those patients.

Dosimetric evaluation of LINAC-based single-isocenter multi-target multi-fraction stereotactic radiosurgery with more than 20 targets: comparing MME, HyperArc, and RapidArc.

BACKGROUND: To compare the dosimetric quality of three widely used techniques for LINAC-based single-isocenter multi-target multi-fraction stereotactic radiosurgery (fSRS) with more than 20 targets: dynamic conformal arc (DCA) in BrainLAB Multiple Metastases Elements (MME) module and volumetric modulated arc therapy (VMAT) using RapidArc (RA) and HyperArc (HA) in Varian Eclipse. METHODS: Ten patients who received single-isocenter fSRS with 20-37 targets were retrospectively replanned using MME, RA, and HA. Various dosimetric parameters, such as conformity index (CI), Paddick CI, gradient index (GI), normal brain dose exposures, maximum organ-at-risk (OAR) doses, and beam-on times were extracted and compared among the three techniques. Wilcoxon signed-rank test was used for statistical analysis. RESULTS: All plans achieved the prescribed dose coverage goal of at least 95% of the planning target volume (PTV). HA plans showed superior conformity compared to RA and MME plans. MME plans showed superior GI compared to RA and HA plans. RA plans resulted in significantly higher low and intermediate dose exposure to normal brain compared to HA and MME plans, especially for lower doses of ≥ 8Gy and ≥ 5Gy. No significant differences were observed in the maximum dose to OARs among the three techniques. The beam-on time of MME plans was about two times longer than RA and HA plans. CONCLUSIONS: HA plans achieved the best conformity, while MME plans achieved the best dose fall-off for LINAC-based single-isocenter multi-target multi-fraction SRS with more than 20 targets. The choice of the optimal technique should consider the trade-offs between dosimetric quality, beam-on time, and planning effort.

A comprehensive evaluation of advanced dose calculation algorithms for brain stereotactic radiosurgery.

PURPOSE: Accurate dose calculation is important in both target and low dose normal tissue regions for brain stereotactic radiosurgery (SRS). In this study, we aim to evaluate the dosimetric accuracy of the two advanced dose calculation algorithms for brain SRS. METHODS: Retrospective clinical case study and phantom study were performed. For the clinical study, 138 SRS patient plans (443 targets) were generated using BrainLab Elements Voxel Monte Carlo (VMC). To evaluate the dose calculation accuracy, the plans were exported into Eclipse and recalculated with Acuros XB (AXB) algorithm with identical beam parameters. The calculated dose at the target center (Dref), dose to 95% target volume (D95), and the average dose to target (Dmean) were compared. Also, the distance from the skull was analyzed. For the phantom study, a cylindrical phantom and a head phantom were used, and the delivered dose was measured by an ion chamber and EBT3 film, respectively, at various locations. The measurement was compared with the calculated doses from VMC and AXB. RESULTS: In clinical cases, VMC dose calculations tended to be higher than AXB. It was found that the difference in Dref showed > 5% in some cases for smaller volumes < 0.3 cm3 . Dmean and D95 differences were also higher for small targets. No obvious trend was found between the dose difference and the distance from the skull. In phantom studies, VMC dose was also higher than AXB for smaller targets, and VMC showed better agreement with the measurements than AXB for both point dose and high dose spread. CONCLUSION: The two advanced calculation algorithms were extensively compared. For brain SRS, AXB sometimes calculates a noticeable lower target dose for small targets than VMC, and VMC tends to have a slightly closer agreement with measurements than AXB.

Influence of air mapping errors on the dosimetric accuracy of prostate CBCT-guided online adaptive radiation therapy.

PURPOSE: CBCT-guided online adaptive radiotherapy (oART) plans presently utilize daily synthetic CTs (sCT) that are automatically generated using deformable registration algorithms. These algorithms may have poor performance at reproducing variable volumes of gas present during treatment. Therefore, we have analyzed the air mapping error between the daily CBCTs and the corresponding sCT and explored its dosimetric effect on oART plan calculation. METHODS: Abdominopelvic air volume was contoured on both the daily CBCT images and the corresponding synthetic images for 207 online adaptive pelvic treatments. Air mapping errors were tracked over all fractions. For two case studies representing worst case scenarios, dosimetric effects of air mapping errors were corrected in the sCT images using the daily CBCT air contours, then recalculating dose. Dose volume histogram statistics and 3D gamma passing rates were used to compare the original and air-corrected sCT-based dose calculations. RESULTS: All analyzed patients showed observable air pocket contour differences between the sCT and the CBCT images. The largest air volume difference observed in daily CBCT images for a given patient was 276.3 cc, a difference of more than 386% compared to the sCT. For the two case studies, the largest observed change in DVH metrics was a 2.6% reduction in minimum PTV dose, with all other metrics varying by less than 1.5%. 3D gamma passing rates using 1%/1 mm criteria were above 90% when comparing the uncorrected and corrected dose distributions. CONCLUSION: Current CBCT-based oART workflow can lead to inaccuracies in the mapping of abdominopelvic air pockets from daily CBCT to the sCT images used for the optimization and calculation of the adaptive plan. Despite the large observed mapping errors, the dosimetric effects of such differences on the accuracy of the adapted plan dose calculation are unlikely to cause differences greater than 3% for prostate treatments.

Simultaneous Image Reconstruction and Element Decomposition for Iodine Contrast Agent Visualization in Multienergy Element-Resolved Cone Beam CT.

Iodine contrast agent is widely used in liver cancer radiotherapy at CT simulation stage to enhance detectability of tumor. However, its application in cone beam CT (CBCT) for image guidance before treatment delivery is still limited because of poor image quality and excessive dose of contrast agent during multiple treatment fractions. We previously developed a multienergy element-resolved (MEER) CBCT framework that included x-ray projection data acquisition on a conventional CBCT platform in a kVp-switching model and a dictionary-based image reconstruction algorithm that simultaneously reconstructed x-ray attenuation images at each kilovoltage peak (kVp), an electron density image, and elemental composition images. In this study, we investigated feasibility using MEER-CBCT for low-concentration iodine contrast agent visualization. We performed simulation and experimental studies using a phantom with inserts containing water and different concentrations of iodine solution and the MEER-CBCT scan with 600 projections in a full gantry rotation, in which the kVp level sequentially changed among 80, 100, and 120 kVps. We included iodine material in the dictionary of the reconstruction algorithm. We analyzed iodine detectability as quantified by contrast-to-noise ratio (CNR) and compared results with those of CBCT images reconstructed by the standard filter back projection (FBP) method with 600 projections. MEER-CBCT achieved similar contrast enhancement as FBP method but significantly higher CNR. At 2.5% iodine solution concentration, FBP method achieved 170 HU enhancement and CNR of 2.0, considered the standard CNR for successful tumor visualization. MEER-CBCT achieved the same CNR but at ~6.3 times lower iodine concentration of 0.4%.

Development of a time-resolved mirrorless scintillation detector.

PURPOSE: We developed a compact and lightweight time-resolved mirrorless scintillation detector (TRMLSD) employing image processing techniques and a convolutional neural network (CNN) for high-resolution two-dimensional (2D) dosimetry. METHODS: The TRMLSD comprises a camera and an inorganic scintillator plate without a mirror. The camera was installed at a certain angle from the horizontal plane to collect scintillation from the scintillator plate. The geometric distortion due to the absence of a mirror and camera lens was corrected using a projective transform. Variations in brightness due to the distance between the image sensor and each point on the scintillator plate and the inhomogeneity of the material constituting the scintillator were corrected using a 20.0 × 20.0 cm2 radiation field. Hot pixels were removed using a frame-based noise-reduction technique. Finally, a CNN-based 2D dose distribution deconvolution model was applied to compensate for the dose error in the penumbra region and a lack of backscatter. The linearity, reproducibility, dose rate dependency, and dose profile were tested for a 6 MV X-ray beam to verify dosimeter characteristics. Gamma analysis was performed for two simple and 10 clinical intensity-modulated radiation therapy (IMRT) plans. RESULTS: The dose linearity with brightness ranging from 0.0 cGy to 200.0 cGy was 0.9998 (R-squared value), and the root-mean-square error value was 1.010. For five consecutive measurements, the reproducibility was within 3% error, and the dose rate dependency was within 1%. The depth dose distribution and lateral dose profile coincided with the ionization chamber data with a 1% mean error. In 2D dosimetry for IMRT plans, the mean gamma passing rates with a 3%/3 mm gamma criterion for the two simple and ten clinical IMRT plans were 96.77% and 95.75%, respectively. CONCLUSION: The verified accuracy and time-resolved characteristics of the dosimeter may be useful for the quality assurance of machines and patient-specific quality assurance for clinical step-and-shoot IMRT plans.

Semi-automatic sigmoid colon segmentation in CT for radiation therapy treatment planning via an iterative 2.5-D deep learning approach.

Automatic sigmoid colon segmentation in CT for radiotherapy treatment planning is challenging due to complex organ shape, close distances to other organs, and large variations in size, shape, and filling status. The patient bowel is often not evacuated, and CT contrast enhancement is not used, which further increase problem difficulty. Deep learning (DL) has demonstrated its power in many segmentation problems. However, standard 2-D approaches cannot handle the sigmoid segmentation problem due to incomplete geometry information and 3-D approaches often encounters the challenge of a limited training data size. Motivated by human's behavior that segments the sigmoid slice by slice while considering connectivity between adjacent slices, we proposed an iterative 2.5-D DL approach to solve this problem. We constructed a network that took an axial CT slice, the sigmoid mask in this slice, and an adjacent CT slice to segment as input and output the predicted mask on the adjacent slice. We also considered other organ masks as prior information. We trained the iterative network with 50 patient cases using five-fold cross validation. The trained network was repeatedly applied to generate masks slice by slice. The method achieved average Dice similarity coefficients of 0.82 0.06 and 0.88 0.02 in 10 test cases without and with using prior information.

Deep-learning assisted automatic digitization of interstitial needles in 3D CT image based high dose-rate brachytherapy of gynecological cancer.

Digitization of interstitial needles is a complicated and tedious process for the treatment planning of 3D CT image based interstitial high dose-rate brachytherapy (HDRBT) of gynecological cancer. We developed a deep-learning assisted auto-digitization method for interstitial needles. The digitization method consisted of two steps. The first step used a deep neural network with a U-net structure to segment all needles from CT images. The second step simultaneously clustered the segmented voxels into different needle groups and generated the needle central trajectories by solving an optimization problem. We evaluated the effectiveness of the developed method in ten interstitial HDRBT patient cases that were not used in the training of the U-net. Average number of needles per case was 20.7. For the segmentation step, average Dice similarity coefficient between automatic and manual segmentation was 0.93. For the digitization step, Hausdorff distance between needle trajectories determined by our method and manually by qualified medical physicists was ~0.71 mm on average and mean difference of tip positions was ~0.63 mm, which were considered acceptable for HDRBT treatment planning. It took ~5 min to complete the digitization process of an interstitial HDRBT case. The achieved accuracy and efficiency made our method clinically attractive.

Deep-learning-assisted automatic digitization of applicators in 3D CT image-based high-dose-rate brachytherapy of gynecological cancer.

PURPOSE: Applicator digitization is one of the most critical steps in 3D high-dose-rate brachytherapy (HDRBT) treatment planning. Motivated by recent advances in deep-learning, we propose a deep-learning-assisted applicator digitization method for 3D CT image-based HDRBT. This study demonstrates its feasibility and potential in gynecological cancer HDRBT. METHODS AND MATERIALS: Our method consisted of two steps. The first step used a U-net to segment applicator regions. We trained the U-net using two-dimensional CT images with a tandem-and-ovoid (T&O) applicator and corresponding applicator mask images. The second step applied a spectral clustering method and a polynomial curve fitting method to extract applicator central paths. We evaluated the accuracy, efficiency, and robustness of our method in different scenarios including other T&O cases that were not used in training, a T&O case scanned with cone-beam CT, and Y-tandem and cylinder-applicator cases. RESULTS: In test cases with a T&O applicator, average 3D Dice similarity coefficient between automatic and manual segmented applicator regions was 0.93. Average distance between tip positions and average Hausdorff distance between applicator channels determined by our method and manually were 0.64 mm and 0.68 mm, respectively. Although trained only using CT images of T&O cases, our tool can also digitize Y-tandem, cylinder applicator, and T&O applicator scanned in cone-beam CT with error of tip position and Hausdorff distance <1 mm. Computation time was ∼15 s per case. CONCLUSIONS: We have developed a deep-learning-assisted applicator digitization tool for 3D CT image-based HDRBT of gynecological cancer. The achieved accuracy, efficiency, and robustness made our tool clinically attractive.

Intelligent inverse treatment planning via deep reinforcement learning, a proof-of-principle study in high dose-rate brachytherapy for cervical cancer.

Inverse treatment planning in radiation therapy is formulated as solving optimization problems. The objective function and constraints consist of multiple terms designed for different clinical and practical considerations. Weighting factors of these terms are needed to define the optimization problem. While a treatment planning optimization engine can solve the optimization problem with given weights, adjusting the weights to yield a high-quality plan is typically performed by a human planner. Yet the weight-tuning task is labor intensive, time consuming, and it critically affects the final plan quality. An automatic weight-tuning approach is strongly desired. The procedure of weight adjustment to improve the plan quality is essentially a decision-making problem. Motivated by the tremendous success in deep learning for decision making with human-level intelligence, we propose a novel framework to adjust the weights in a human-like manner. This study used inverse treatment planning in high-dose-rate brachytherapy (HDRBT) for cervical cancer as an example. We developed a weight-tuning policy network (WTPN) that observes dose volume histograms of a plan and outputs an action to adjust organ weighting factors, similar to the behaviors of a human planner. We trained the WTPN via end-to-end deep reinforcement learning. Experience replay was performed with the epsilon greedy algorithm. After training was completed, we applied the trained WTPN to guide treatment planning of five testing patient cases. It was found that the trained WTPN successfully learnt the treatment planning goals and was able to guide the weight tuning process. On average, the quality score of plans generated under the WTPN's guidance was improved by ~8.5% compared to the initial plan with arbitrarily set weights, and by 10.7% compared to the plans generated by human planners. To our knowledge, this was the first time that a tool was developed to adjust organ weights for the treatment planning optimization problem in a human-like fashion based on intelligence learnt from a training process, which was different from existing strategies based on pre-defined rules. The study demonstrated potential feasibility to develop intelligent treatment planning approaches via deep reinforcement learning.

Development of a real-time internal and external marker tracking system for particle therapy: a phantom study using patient tumor trajectory data.

Target motion-induced uncertainty in particle therapy is more complicated than that in X-ray therapy, requiring more accurate motion management. Therefore, a hybrid motion-tracking system that can track internal tumor motion and as well as an external surrogate of tumor motion was developed. Recently, many correlation tests between internal and external markers in X-ray therapy have been developed; however, the accuracy of such internal/external marker tracking systems, especially in particle therapy, has not yet been sufficiently tested. In this article, the process of installing an in-house hybrid internal/external motion-tracking system is described and the accuracy level of tracking system was acquired. Our results demonstrated that the developed in-house external/internal combined tracking system has submillimeter accuracy, and can be clinically used as a particle therapy system as well as a simulation system for moving tumor treatment.

A virtual simulator designed for collision prevention in proton therapy.

PURPOSE: In proton therapy, collisions between the patient and nozzle potentially occur because of the large nozzle structure and efforts to minimize the air gap. Thus, software was developed to predict such collisions between the nozzle and patient using treatment virtual simulation. METHODS: Three-dimensional (3D) modeling of a gantry inner-floor, nozzle, and robotic-couch was performed using SolidWorks based on the manufacturer's machine data. To obtain patient body information, a 3D-scanner was utilized right before CT scanning. Using the acquired images, a 3D-image of the patient's body contour was reconstructed. The accuracy of the image was confirmed against the CT image of a humanoid phantom. The machine components and the virtual patient were combined on the treatment-room coordinate system, resulting in a virtual simulator. The simulator simulated the motion of its components such as rotation and translation of the gantry, nozzle, and couch in real scale. A collision, if any, was examined both in static and dynamic modes. The static mode assessed collisions only at fixed positions of the machine's components, while the dynamic mode operated any time a component was in motion. A collision was identified if any voxels of two components, e.g., the nozzle and the patient or couch, overlapped when calculating volume locations. The event and collision point were visualized, and collision volumes were reported. RESULTS: All components were successfully assembled, and the motions were accurately controlled. The 3D-shape of the phantom agreed with CT images within a deviation of 2 mm. Collision situations were simulated within minutes, and the results were displayed and reported. CONCLUSIONS: The developed software will be useful in improving patient safety and clinical efficiency of proton therapy.