Treatment Planning Methods for Dose Painting by Numbers Treatment in Gamma Knife Radiosurgery

Purpose

Dose painting radiation therapy delivers a nonuniform dose to tumors to account for heterogeneous radiosensitivity. With recent and ongoing development of Gamma Knife machines making large-volume brain tumor treatments more practical, it is increasingly feasible to deliver dose painting treatments. The increased prescription complexity means automated treatment planning is greatly beneficial, and the impact of dose painting on stereotactic radiosurgery (SRS) plan quality has not yet been studied. This research investigates the plan quality achievable for Gamma Knife SRS dose painting treatments when using optimization techniques and automated isocenter placement in treatment planning.

Methods and Materials

Dose painting prescription functions with varying parameters were applied to convert voxel image intensities to prescriptions for 10 sample cases. To study achievable plan quality and optimization, clinically placed isocenters were used with each dose painting prescription and optimized using a semi-infinite linear programming formulation. To study automated isocenter placement, a grassfire sphere-packing algorithm and a clinically available Leksell gamma plan isocenter fill algorithm were used. Plan quality for each optimized treatment plan was measured with dose painting SRS metrics.

Results

Optimization can be used to find high quality dose painting plans, and plan quality is affected by the dose painting prescription method. Polynomial function prescriptions show more achievable plan quality than sigmoid function prescriptions even with high mean dose boost. Automated isocenter placement is shown as a feasible method for dose painting SRS treatment, and increasing the number of isocenters improves plan quality. The computational solve time for optimization is within 5 minutes in most cases, which is suitable for clinical planning.

Conclusions

The impact of dose painting prescription method on achievable plan quality is quantified in this study. Optimization and automated isocenter placement are shown as possible treatment planning methods to obtain high quality plans.

Automatic Inverse Treatment Planning of Gamma Knife Radiosurgery via Deep Reinforcement Learning

PURPOSE

Several inverse planning algorithms have been developed for Gamma Knife (GK) radiosurgery to determine a large number of plan parameters via solving an optimization problem, which typically consists of multiple objectives. The priorities among these objectives need to be repetitively adjusted to achieve a clinically good plan for each patient. This study aimed to achieve automatic and intelligent priority-tuning, by developing a deep reinforcement learning (DRL) based method to model the tuning behaviors of human planners.

METHODS

We built a priority-tuning policy network using deep convolutional neural networks. Its input was a vector composed of multiple plan metrics that were used in our institution for GK plan evaluation. The network can determine which tuning action to take, based on the observed quality of the intermediate plan. We trained the network using an end-to-end DRL framework to approximate the optimal action-value function. A scoring function was designed to measure the plan quality to calculate the received reward of a tuning action.

RESULTS

Vestibular schwannoma was chosen as the test bed in this study. The number of training, validation and testing cases were 5, 5, and 16, respectively. For these three datasets, the average scores of the initial plans obtained with a same initial priority set were 3.63 ± 1.34, 3.83 ± 0.86 and 4.20 ± 0.78, respectively, while can be improved to 5.28 ± 0.23, 4.97 ± 0.44 and 5.22 ± 0.26 through manual priority tuning by human expert planners. Our network achieved competitive results with 5.42 ± 0.11, 5.10 ± 0. 42, 5.28 ± 0.20, respectively.

CONCLUSIONS

Our network can generate GK plans of comparable or slightly higher quality comparing with the plans generated by human planners via manual priority tuning for vestibular schwannoma cases. The network can potentially be incorporated into the clinical workflow as a planning assistance to improve GK planning efficiency and help to reduce plan quality variation caused by inter-planner variability. We also hope that our method can reduce the workload of GK planners and allow them to spend more time on more challenging cases. This article is protected by copyright. All rights reserved.

Dosimetry

The dosimeters used to measure radiation dose produce a value which has to be calibrated to be in keeping with the values in an approved laboratory, which will be one of an international network of such laboratories at the center of which is the Bureau International des Poids et Mesures in France (BIPM). Dosimeters work by producing a quantitatively proportional change in status to the intensity of the radiation being measure. Amongst the techniques in use are thermoluminescent devices, radiographic film, radiochromic film, semiconductors, ionization chambers, silicon diodes and gel dosimeters. The Gamma Knife radiation has been difficult to measure directly because the beams have been to fine for accurate measurement by commonly available dosimeters. For more modern dosimeters this is less of a problem. During the treatment of a patient, a variety of indices are recorded to assist in the standardization and accuracy of treatment. Having determined the dose in the beams, it is necessary to calculate how much energy is lost during the passage of radiation from the source to the target. There has been a steady evolution of these calculations to make them more accurate.

Inverse planning in Gamma Knife radiosurgery: A comparative planning study

PURPOSE

To determine the advantages of inverse planning using a prerelease version of Leksell Gamma Knife® (LGK) Lightning (Elekta AB, Sweden) compared to manual forward planning.

METHODS

Thirty-eight patients with metastases (MET, n = 15), vestibular schwannomas (VS, n = 11) and meningiomas (MEN, n = 12), treated with LGK Icon™ at our institution, were analyzed retrospectively. For each case, an inverse (inv) and inverse full coverage (fc) treatment plan was generated using LGK Lightning and compared to the clinical plans. Several dosimetry and efficiency characteristics were compared for each indication. The mean, median difference and interquartile range were reported and the significance was assessed with a paired-sample Wilcoxon test (significance level < 0.05). Further, the inter operator variability was analyzed for multiple users.

RESULTS

Inv and fc treatment plans show improved target coverage (up to 3.6%) for all analyzed paradigms. For inv plans, the selectivity is enhanced (MET: 2.9%; VS: 1.8%; MEN: 1%) and the organ at risk doses are significantly reduced (VS: up to 4.5%; MEN: up to 17.5%). For inv and fc plans, the beam on time (BOT) is shortened (MET: up to 7.9%; benign tumors: 49.5%). The inter operator variability analysis shows similar treatment plan quality with small differences in plan efficiency (difference in BOT: 1-3.3 min).

CONCLUSIONS

LGK Lightning allows to generate improved LGK treatment plans regarding plan quality with reduced BOT compared to manual forward plans. The inter operator variability showed that multiple users with different experiences can generate similar treatment plan quality using LGK Lightning.

A linear programming approach to inverse planning in Gamma Knife radiosurgery

PURPOSE

Leksell Gamma Knife® is a stereotactic radiosurgery system that allows fine-grained control of the delivered dose distribution. We describe a new inverse planning approach that both resolves shortcomings of earlier approaches and unlocks new capabilities.

METHODS

We fix the isocenter positions and perform sector-duration optimization using linear programming, and study the effect of beam-on time penalization on the trade-off between beam-on time and plan quality. We also describe two techniques that reduce the problem size and thus further reduce the solution time: dualization and representative subsampling.

RESULTS

The beam-on time penalization reduces the beam-on time by a factor 2-3 compared with the naïve alternative. Dualization and representative subsampling each leads to optimization time-savings by a factor 5-20. Overall, we find in a comparison with 75 clinical plans that we can always find plans with similar coverage and better selectivity and beam-on time. In 44 of these, we can even find a plan that also has better gradient index. On a standard GammaPlan workstation, the optimization times ranged from 2.3 to 26 s with a median time of 5.7 s.

CONCLUSION

We present a combination of techniques that enables sector-duration optimization in a clinically feasible time frame.

Simultaneous optimization of isocenter locations and sector duration in radiosurgery

Stereotactic radiosurgery is an effective technique to treat brain tumors for which several inverse planning methods may be appropriate. We propose an integer programming model to simultaneous sector duration and isocenter optimization (SDIO) problem for Leksell Gamma Knife^® Icon^TM (Elekta, Stockholm, Sweden) to tractably incorporate treatment time. We devise a Benders decomposition scheme to solve the SDIO problem to optimality. The performances of our approaches are assessed using anonymized data from eight previously treated cases, and obtained treatment plans are compared against each other and against the clinical plans. The plans generated by our SDIO model all meet or exceed clinical guidelines while demonstrating high conformity.

Modeling and comparison of alternative approaches for sector duration optimization in a dedicated radiosurgery system

Stereotactic radiosurgery (SRS) is an effective technique to treat brain metastasis for which several inverse planning methods may be appropriate. We compare three different optimization models for segment duration optimization in SRS using Leksell Gamma Knife^® Icon^TM (Elekta, Stockholm, Sweden). We investigate (1) a linear programming approach, (2) a piecewise quadratic penalty approach, and (3) an unconstrained convex moment-based penalty approach. We examine the performances of these approaches using anonymized data from 14 previously treated cases. In addition, we investigate the important modeling question of selecting weights for the objective functions where we use a simulated annealing algorithm to determine these weights for each model. The inverse plans obtained via optimization models are compared against each other and against the clinical plans. The three inverse planning models can all yield optimal treatment plans in a reasonable amount of time and the treatment plans obtained by these models meet or exceed clinical guidelines while displaying high conformity.

The Use of Cone Beam Computed Tomography for Image Guided Gamma Knife Stereotactic Radiosurgery: Initial Clinical Evaluation

PURPOSE

The present study used cone beam computed tomography (CBCT) to measure the inter- and intrafraction uncertainties for intracranial stereotactic radiosurgery (SRS) using the Leksell Gamma Knife (GK).

METHODS AND MATERIALS

Using a novel CBCT system adapted to the GK radiosurgery treatment unit, CBCT images were acquired immediately before and after treatment for each treatment session within the context of a research ethics board-approved prospective clinical trial. Patients were immobilized in the Leksell coordinate frame (LCF) for both volumetric CBCT imaging and GK-SRS delivery. The relative displacement of the patient's skull to the stereotactic reference (interfraction motion) was measured for each CBCT scan. Differences between the pre- and post-treatment CBCT scans were used to determine the intrafraction motion.

RESULTS

We analyzed 20 pre- and 17 post-treatment CBCT scans in 20 LCF patients treated with SRS. The mean translational pretreatment setup error ± standard deviation in the left-right, anteroposterior, and craniocaudal directions was -0.19 ± 0.32, 0.06 ± 0.27, and -0.23 ± 0.2 mm, with a maximum of -0.74, -0.53, and -0.68 mm, respectively. After an average time between the pre- and post-treatment CBCT scans of 82 minutes (range 27-170), the mean intrafraction error ± standard deviation for the LCF was -0.03 ± 0.05, -0.03 ± 0.18, and -0.03 ± 0.12 mm in the left-right, anteroposterior, and craniocaudual direction, respectively.

CONCLUSIONS

Using CBCT on a prototype image guided GK Perfexion unit, we were able to measure the inter- and intrafraction positional changes for GK-SRS using the invasive frame. In the era of image guided radiation therapy, the use of CBCT image guidance for both frame- and non-frame-based immobilization systems could serve as a useful quality assurance tool. Our preliminary measurements can guide the application of achievable thresholds for inter- and intrafraction discrepancy when moving to a frameless approach.

Knowledge-based prediction of plan quality metrics in intracranial stereotactic radiosurgery

PURPOSE

The objective of this work was to develop a comprehensive knowledge-based methodology for predicting achievable dose-volume histograms (DVHs) and highly precise DVH-based quality metrics (QMs) in stereotactic radiosurgery/radiotherapy (SRS/SRT) plans. Accurate QM estimation can identify suboptimal treatment plans and provide target optimization objectives to standardize and improve treatment planning.

METHODS

Correlating observed dose as it relates to the geometric relationship of organs-at-risk (OARs) to planning target volumes (PTVs) yields mathematical models to predict achievable DVHs. In SRS, DVH-based QMs such as brain V10Gy (volume receiving 10 Gy or more), gradient measure (GM), and conformity index (CI) are used to evaluate plan quality. This study encompasses 223 linear accelerator-based SRS/SRT treatment plans (SRS plans) using volumetric-modulated arc therapy (VMAT), representing 95% of the institution's VMAT radiosurgery load from the past four and a half years. Unfiltered models that use all available plans for the model training were built for each category with a stratification scheme based on target and OAR characteristics determined emergently through initial modeling process. Model predictive accuracy is measured by the mean and standard deviation of the difference between clinical and predicted QMs, δQM = QMclin - QMpred, and a coefficient of determination, R(2). For categories with a large number of plans, refined models are constructed by automatic elimination of suspected suboptimal plans from the training set. Using the refined model as a presumed achievable standard, potentially suboptimal plans are identified. Predictions of QM improvement are validated via standardized replanning of 20 suspected suboptimal plans based on dosimetric predictions. The significance of the QM improvement is evaluated using the Wilcoxon signed rank test.

RESULTS

The most accurate predictions are obtained when plans are stratified based on proximity to OARs and their PTV volume sizes. Volumes are categorized into small (VPTV < 2 cm(3)), medium (2 cm(3) < VPTV < 25 cm(3)), and large (25 cm(3) < VPTV). The unfiltered models demonstrate the ability to predict GMs to ∼1 mm and fractional brain V10Gy to ∼25% for plans with large VPTV and critical OAR involvements. Increased accuracy and precision of QM predictions are obtained when high quality plans are selected for the model training. For the small and medium VPTV plans without critical OAR involvement, predictive ability was evaluated using the refined model. For training plans, the model predicted GM to an accuracy of 0.2 ± 0.3 mm and fractional brain V10Gy to 0.04 ± 0.12, suggesting highly accurate predictive ability. For excluded plans, the average δGM was 1.1 mm and fractional brain V10Gy was 0.20. These δQM are significantly greater than those of the model training plans (p < 0.001). For CI, predictions are close to clinical values and no significant difference was observed between the training and excluded plans (p = 0.19). Twenty outliers with δGM > 1.35 mm were identified as potentially suboptimal, and replanning these cases using predicted target objectives demonstrates significant improvements on QMs: on average, 1.1 mm reduction in GM (p < 0.001) and 23% reduction in brain V10Gy (p < 0.001). After replanning, the difference of δGM distribution between the 20 replans and the refined model training plans was marginal.

CONCLUSIONS

The results demonstrate the ability to predict SRS QMs precisely and to identify suboptimal plans. Furthermore, the knowledge-based DVH predictions were directly used as target optimization objectives and allowed a standardized planning process that bettered the clinically approved plans. Full clinical application of this methodology can improve consistency of SRS plan quality in a wide range of PTV volume and proximity to OARs and facilitate automated treatment planning for this critical treatment site.