Y Sheng

Standardized beam bouquets for lung IMRT planning

The selection of the incident angles of the treatment beams is a critical component of intensity modulated radiation therapy (IMRT) planning for lung cancer due to significant variations in tumor location, tumor size and patient anatomy. We investigate the feasibility of establishing a small set of standardized beam bouquets for planning. The set of beam bouquets were determined by learning the beam configuration features from 60 clinical lung IMRT plans designed by experienced planners. A k-medoids cluster analysis method was used to classify the beam configurations in the dataset. The appropriate number of clusters was determined by maximizing the value of average silhouette width of the classification. Once the number of clusters had been determined, the beam arrangements in each medoid of the clusters were designated as the standardized beam bouquet for the cluster. This standardized bouquet set was used to re-plan 20 cases randomly selected from the clinical database. The dosimetric quality of the plans using the beam bouquets was evaluated against the corresponding clinical plans by a paired t-test. The classification with six clusters has the largest average silhouette width value and hence would best represent the beam bouquet patterns in the dataset. The results shows that plans generated with a small number of standardized bouquets (e.g. 6) have comparable quality to that of clinical plans. These standardized beam configuration bouquets will potentially help improve plan efficiency and facilitate automated planning.

Atlas-guided prostate intensity modulated radiation therapy (IMRT) planning.

An atlas-based IMRT planning technique for prostate cancer was developed and evaluated. A multi-dose atlas was built based on the anatomy patterns of the patients, more specifically, the percent distance to the prostate and the concaveness angle formed by the seminal vesicles relative to the anterior-posterior axis. A 70-case dataset was classified using a k-medoids clustering analysis to recognize anatomy pattern variations in the dataset. The best classification, defined by the number of classes or medoids, was determined by the largest value of the average silhouette width. Reference plans from each class formed a multi-dose atlas. The atlas-guided planning (AGP) technique started with matching the new case anatomy pattern to one of the reference cases in the atlas; then a deformable registration between the atlas and new case anatomies transferred the dose from the atlas to the new case to guide inverse planning with full automation. 20 additional clinical cases were re-planned to evaluate the AGP technique. Dosimetric properties between AGP and clinical plans were evaluated. The classification analysis determined that the 5-case atlas would best represent anatomy patterns for the patient cohort. AGP took approximately 1 min on average (corresponding to 70 iterations of optimization) for all cases. When dosimetric parameters were compared, the differences between AGP and clinical plans were less than 3.5%, albeit some statistical significances observed: homogeneity index (p  >  0.05), conformity index (p  <  0.01), bladder gEUD (p  <  0.01), and rectum gEUD (p  =  0.02). Atlas-guided treatment planning is feasible and efficient. Atlas predicted dose can effectively guide the optimizer to achieve plan quality comparable to that of clinical plans.

Outlier identification in radiation therapy knowledge‐based planning: A study of pelvic cases

Purpose: The purpose of this study was to apply statistical metrics to identify outliers and to investigate the impact of outliers on knowledge‐based planning in radiation therapy of pelvic cases. We also aimed to develop a systematic workflow for identifying and analyzing geometric and dosimetric outliers. Methods: Four groups (G1–G4) of pelvic plans were sampled in this study. These include the following three groups of clinical IMRT cases: G1 (37 prostate cases), G2 (37 prostate plus lymph node cases) and G3 (37 prostate bed cases). Cases in G4 were planned in accordance with dynamic‐arc radiation therapy procedure and include 10 prostate cases in addition to those from G1. The workflow was separated into two parts: 1. identifying geometric outliers, assessing outlier impact, and outlier cleaning; 2. identifying dosimetric outliers, assessing outlier impact, and outlier cleaning. G2 and G3 were used to analyze the effects of geometric outliers (first experiment outlined below) while G1 and G4 were used to analyze the effects of dosimetric outliers (second experiment outlined below).A baseline model was trained by regarding all G2 cases as inliers. G3 cases were then individually added to the baseline model as geometric outliers. The impact on the model was assessed by comparing leverages of inliers (G2) and outliers (G3). A receiver‐operating‐characteristic (ROC) analysis was performed to determine the optimal threshold. The experiment was repeated by training the baseline model with all G3 cases as inliers and perturbing the model with G2 cases as outliers.A separate baseline model was trained with 32 G1 cases. Each G4 case (dosimetric outlier) was subsequently added to perturb the model. Predictions of dose‐volume histograms (DVHs) were made using these perturbed models for the remaining 5 G1 cases. A Weighted Sum of Absolute Residuals (WSAR) was used to evaluate the impact of the dosimetric outliers. Results: The leverage of inliers and outliers was significantly different. The Area‐Under‐Curve (AUC) for differentiating G2 (outliers) from G3 (inliers) was 0.98 (threshold: 0.27) for the bladder and 0.81 (threshold: 0.11) for the rectum. For differentiating G3 (outlier) from G2 (inlier), the AUC (threshold) was 0.86 (0.11) for the bladder and 0.71 (0.11) for the rectum. Significant increase in WSAR was observed in the model with 3 dosimetric outliers for the bladder (P < 0.005 with Bonferroni correction), and in the model with only 1 dosimetric outlier for the rectum (P < 0.005). Conclusions: We established a systematic workflow for identifying and analyzing geometric and dosimetric outliers, and investigated statistical metrics for outlier detection. Results validated the necessity for outlier detection and clean‐up to enhance model quality in clinical practice.

SU‐E‐T‐537: Local Multi‐Criteria Optimization for Clinical Tradeoff Decision Guidance in RT Planning

Purpose: We present a novel method for generating a local segment of Pareto surface around the best achievable plan predicted by a knowledge model. This local-MCO will provide an efficient method to enable clinically viable tradeoff decisions in IMRT planning tailored to the patient’s specific needs. Methods: Multi-criteria optimization (MCO) methods provide physicians and planners the ability to explore tradeoff options in RT planning. However, generic MCO methods are often time consuming because a global Pareto surface (PS) need to be explored if patient-specific clinical conditions and planners’ planning experiences are not taken into consideration. We have developed a local-MCO approach, which incorporates the knowledge model prediction based on the individual patient’s features and planning experience into the generation of a local PS around a predicted plan. In the proposed method, the starting points of clinical relevant organ sparing objectives are predicted by the knowledge models, and then a local PS is searched near the model predictions. As an initial assessment, the local PS is compared with the global PS generated for two prostate cancer cases. The mean minimum distance from each plan on the local PS to the global PS and the range of clinical acceptable dosimetric parameters covered by the local and global PS are calculated. Results: The local PS agrees well with the global PS. The mean minimum distance between the local and global PS in the PTV-bladder-rectum dose objective space are about 3% and 1% of prescription dose for the two plans, respectively. Although the local PS is only a small portion of the global PS, they cover most of the clinically relevant dose range. Conclusion: The local MCO results in a smaller but clinically more relevant PS. It is an efficient method to provide physicians with guidance of patient-specific trade-off options based on practice experience. Partially supported by NIH/NCI under grant #R21CA161389 and a master research grant by Varian Medical Systems.

Lung IMRT planning with automatic determination of beam angle configurations

Beam angle configuration is a major planning decision in intensity modulated radiation treatment (IMRT) that has a significant impact on dose distributions and thus quality of treatment, especially in complex planning cases such as those for lung cancer treatment. We propose a novel method to automatically determine beam configurations that incorporates noncoplanar beams. We then present a completely automated IMRT planning algorithm that combines the proposed method with a previously reported OAR DVH prediction model. Finally, we validate this completely automatic planning algorithm using a set of challenging lung IMRT cases. A beam efficiency index map is constructed to guide the selection of beam angles. This index takes into account both the dose contributions from individual beams and the combined effect of multiple beams by introducing a beam-spread term. The effect of the beam-spread term on plan quality was studied systematically and the weight of the term to balance PTV dose conformity against OAR avoidance was determined. For validation, complex lung cases with clinical IMRT plans that required the use of one or more noncoplanar beams were re-planned with the proposed automatic planning algorithm. Important dose metrics for the PTV and OARs in the automatic plans were compared with those of the clinical plans. The results are very encouraging. The PTV dose conformity and homogeneity in the automatic plans improved significantly. And all the dose metrics of the automatic plans, except the lung V5 Gy, were statistically better than or comparable with those of the clinical plans. In conclusion, the automatic planning algorithm can incorporate non-coplanar beam configurations in challenging lung cases and can generate plans efficiently with quality closely approximating that of clinical plans.

Temporal lobe injury patterns following intensity modulated radiotherapy in a large cohort of nasopharyngeal carcinoma patients

OBJECTIVES To analyze the correlation between dose-volume-histograms (DVHs) with three patterns (edema, enhancement, and necrosis) of temporal lobe injury (TLI) in patients receiving intensity modulated radiation therapy (IMRT) for nasopharyngeal carcinoma (NPC) and to determine optimal thresholds to predict the incidence of each TLI pattern, with particular emphasis on the relationship between edema volume and the risk of enhancement and necrosis. MATERIALS AND METHODS A cohort of 4186 NPC patients treated with IMRT was retrospectively reviewed with TLI presenting in 188 patients. The atlases of complication incidence (ACI) for each pattern were constructed using DVH curves of temporal lobes. Optimal threshold for predicting incidence of each pattern was determined using the point closest to top-left of the plot. The accuracy of using edema volume to predict enhancement and necrosis incidence was evaluated via area under curve (AUC) of receiver operator characteristics (ROC). RESULTS All DVH parameters, Dmean, Dmax, D0.25cc, D0.5cc, D1cc, D3cc, D6cc, V20Gy, V30Gy, V40Gy, V50Gy, V60Gy, and V70Gy, except Dmin showed statistically significant differences between subgroups of each pattern (p < 0.05). For predicting incidence of each pattern, optimal DVH thresholds over the range of D0.25-D1cc, Dmean and V20-V70 were derived. The optimal thresholds of edema volume for predicting enhancement were 0.96 and 2.2cc and for predicting necrosis were 0.94 and 11.5cc. CONCLUSION Optimal DVH thresholds were generated for limiting risk of each injury pattern. Edema volume was a strong predictor for risk of enhancement and necrosis, which could potentially be reduced by lowering edema volume below threshold.

An Ensemble Approach to Knowledge-Based Intensity-Modulated Radiation Therapy Planning

Knowledge-based planning (KBP) utilizes experienced planners’ knowledge embedded in prior plans to estimate optimal achievable dose volume histogram (DVH) of new cases. In the regression-based KBP framework, previously planned patients’ anatomical features and DVHs are extracted, and prior knowledge is summarized as the regression coefficients that transform features to organ-at-risk DVH predictions. In our study, we find that in different settings, different regression methods work better. To improve the robustness of KBP models, we propose an ensemble method that combines the strengths of various linear regression models, including stepwise, lasso, elastic net, and ridge regression. In the ensemble approach, we first obtain individual model prediction metadata using in-training-set leave-one-out cross validation. A constrained optimization is subsequently performed to decide individual model weights. The metadata is also used to filter out impactful training set outliers. We evaluate our method on a fresh set of retrospectively retrieved anonymized prostate intensity-modulated radiation therapy (IMRT) cases and head and neck IMRT cases. The proposed approach is more robust against small training set size, wrongly labeled cases, and dosimetric inferior plans, compared with other individual models. In summary, we believe the improved robustness makes the proposed method more suitable for clinical settings than individual models.

SU-F-T-341: Generate Clinical Acceptable Trade-Off Options in Brain IMRT Planning by Local Multi-Criteria Optimization (MCO) Method

PURPOSE we present a method to generate a set of treatment plans with clinical manageable dosimetric tradeoff options in brain IMRT planning guided by knowledge model prediction. METHODS Multi-criteria optimization (MCO) methods have been developed in RT planning to help physicians make the complex clinical tradeoff decision among different organ sparing goals. In local MCO method, a clinical acceptable tradeoff range is first predicted based on past planning experiences and the specific patients anatomy. The anchor points of the local Pareto surface (PS) are obtained from the model prediction. The Pareto front is further refined by varying the objective within the acceptable trade-off range in plan optimization to generate additional point between the anchor points. The quality and efficiency of this method are validated by brain IMRT planning in this study. The knowledge model was first trained by 15 clinical brain IMRT plans. Then it is utilized to guide the Pareto front search for another set of 5 patient cases. We studied the number of plan optimizations needed in order to generate the Pareto surface with sufficient precision which is indicated by the mean minimum distance between the local PS and the global PS. RESULTS Local PS can be generated efficiently within the clinical tradeoff range. Using only two anchor plans, the mean minimum distance between the local and global PS in the PTV homogeneity-brainstem median dose (D50) objective space for the 5 validation cases are: ∼0.6% to 2% and 0.3% to 1% of prescription dose in terms of PTV dose homogeneity and brainstem D50, respectively. The tradeoff ranges covered by the local PS vary with patient anatomy, which are 10-20% and 40-80% on average for the two dosimetric parameters. CONCLUSION By combining knowledge model and MCO, the local MCO method can generate clinical optimal tradeoff options efficiently.

WE-AB-209-05: Development of an Ultra-Fast High Quality Whole Breast Radiotherapy Treatment Planning System

PURPOSE To enable near-real-time (<20sec) and interactive planning without compromising quality for whole breast RT treatment planning using tangential fields. METHODS Whole breast RT plans from 20 patients treated with single energy (SE, 6MV, 10 patients) or mixed energy (ME, 6/15MV, 10 patients) were randomly selected for model training. Additional 20 cases were used as validation cohort. The planning process for a new case consists of three fully automated steps:1. Energy Selection. A classification model automatically selects energy level. To build the energy selection model, principle component analysis (PCA) was applied to the digital reconstructed radiographs (DRRs) of training cases to extract anatomy-energy relationship.2. Fluence Estimation. Once energy is selected, a random forest (RF) model generates the initial fluence. This model summarizes the relationship between patient anatomys shape based features and the output fluence. 3. Fluence Fine-tuning. This step balances the overall dose contribution throughout the whole breast tissue by automatically selecting reference points and applying centrality correction. Fine-tuning works at beamlet-level until the dose distribution meets clinical objectives. Prior to finalization, physicians can also make patient-specific trade-offs between target coverage and high-dose volumes.The proposed method was validated by comparing auto-plans with manually generated clinical-plans using Wilcoxon Signed-Rank test. RESULTS In 19/20 cases the model suggested the same energy combination as clinical-plans. The target volume coverage V100% was 78.1±4.7% for auto-plans, and 79.3±4.8% for clinical-plans (p=0.12). Volumes receiving 105% Rx were 69.2±78.0cc for auto-plans compared to 83.9±87.2cc for clinical-plans (p=0.13). The mean V10Gy, V20Gy of the ipsilateral lung was 24.4±6.7%, 18.6±6.0% for auto plans and 24.6±6.7%, 18.9±6.1% for clinical-plans (p=0.04, <0.001). Total computational time for auto-plans was < 20s. CONCLUSION We developed an automated method that generates breast radiotherapy plans with accurate energy selection, similar target volume coverage, reduced hotspot volumes, and significant reduction in planning time, allowing for near-real-time planning.

Knowledge based Automatic Beam Angle Determination for Lung IMRT Planning

We present an efficient knowledge based automatic beam configuration determination method by utilizing patient-specific anatomy and tumor geometry information and a beam bouquet atlas. The proposed technique is based on learning the relationship between patient anatomy and beam configurations from clinical plans designed by experienced planners. The training dataset contains 60 lung IMRT plans with plan prescription dose between 45 and 70 Gy. The method involves three major steps. First, a beam bouquet atlas was established by classifying the clinical plans into 6 beam configuration groups using the k-medoids cluster analysis method. Second, a beam efficiency map was constructed to characterize the geometry of the tumor relative to the lungs, the body and the other OARs at each candidate beam direction. Finally, the beam efficiency maps of the clinical cases and the cluster assignments of their beam bouquets were paired to train a Bayesian classification model. This classification model was used to select a suitable beam bouquet from the atlas for a new case based on its beam efficiency map. This technique was validated by leave-one-out cross validation with 16 cases randomly selected from the original dataset. The dosimetric parameters (mean±S.D. in percentage of prescription dose) in the auto-beam plans and in the clinical plans, respectively, and the p-values by a paired t-test (in parenthesis) are: lung mean dose (Dmean): 16.3±9.3, 18.6±7.4 (0.48), esophagus Dmean: 28.4±18, 30.7±19.3 (0.02), Heart Dmean: 21.5±17.5,21.1±17.2 (0.76), Spinal Cord D2%: 48±23, 51.2±21.8 (0.01), PTV dose homogeneity (D2%-D99%): 22±27.4, 20.4±12.8 (0.10). The other dosimetric parameters are not statistically different. In conclusion, plans generated by the automatic beam angle determination method can achieve dosimetric quality equivalent to that of clinical plans. This method can help improve the quality and efficiency of lung IMRT planning.