Knowledge-based Treatment Planning Research Articles

Multi-planner validation of RapidPlan knowledge-based model for volumetric modulated arc therapy in prostate cancer.

To investigate the performance of a model-based optimization process for volumetric modulated arc therapy (VMAT) applied to prostate cancer patients with the multi-planner. The 120 prostate plans for VMAT treatment were entered into the database system of the RapidPlan (RP) knowledge-based treatment planning. The treatment planning data for each plan was used to create and train the RP model. Twelve prostate cancer cases were selected and were used for planning by a manual of 12 planners based on the clinical protocol for dose constraints. Then, the treatment plans for each patient were compared with the RP model plans and analyzed with Wilcoxon tests. On average, the RP models can estimate comparable doses among all planner plans and clinical plans for the PTV, which Dmax , D95% , D98% , HI, and CI were used to evaluate. For the normal organ doses of the bladder, rectum, penile bulb, and femoral head, all RP model plans showed comparable or better dose sparing than all planner plans and clinical plans. Moreover, the average planning time of the RP model was faster than manual plans by about two times. The RP model can significantly reduce the variation dose of the normal organs compared with the manual plans among the planners. The automated plans of the RP model might benefit from further fine-tuning of the dose constraints of the normal organs, although both procedure plans are acceptable and fulfill the clinical protocol goals so that the RP model can enhance the efficacy and quality of plans.

Optimizing Radiation Therapy for Brain Metastasis Patients: A Comparative Study of Knowledge-Based Treatment Planning with HyperArc

The purpose of this study is to examine the benefits of using a knowledge-based treatment planning (KBP) model in treating brain metastasis patients with HyperArc (HA) automated non-coplanar stereotactic radiosurgery (SRS). Treatment plans of the 113 patients with either single or multiple brain metastases treated with HA were enrolled in the study. The eighty-two treatment plans with 190 metastatic brain lesions were used to construct the training model using the DVH estimation algorithm. The model was then validated on the remaining 31 patients' treatment plans with 56 metastatic brain lesions. Two corresponding HA plans were manually generated for these 31 patients using the two optimization algorithms (SRS NTO and ALDO). The SRS NTO plan generates a rapid dose falloff and prevents dose bridging of adjacent disparate targets, while the ALDO plan can cover the target with the prescribed dose but may create a high inhomogeneous dose within the target. The dosimetric outcomes were compared between the KBP, SRS NTO, and ALDO plans. The KBP, SRS NTO, and ALDO plan consistently showed CTV and PTV coverage levels of over 99%. The KBP plans resulted in similar doses to the organs at risk compared to the SRS NTO plans, but significantly reduced doses to the brainstem and optic apparatus (lens, eyes, optic nerves, and optic chiasma) by 17% to 67% compared to the ALDO plans (P = 0.003 for brainstem and P < 0.001 for optic apparatus). The mean brain dose was not significantly different between the three plans (P = 0.694 for KBP vs. SRS NTO and P = 0.381 for KBP vs. ALDO). However, the ALDO plans showed a statistically better brain V12 Gy than the KBP plans (P = 0.009), with a difference of less than 2 c.c. that was deemed not clinically significant. The ALDO plans also had a significantly better intermediate dose spillage, gradient radius, and conformity index than the KBP plans. The KBP plans showed a significantly faster gradient index and intermediate dose spillage compared to the SRS NTO plans. This study demonstrates the ability of the KBP model (RapidPlan) to generate accurate DVH predictions for use with HA in generating SRS plans that achieve optimal sparing of organs at risk while maintaining dose falloff. The KBP plans showed balanced dosimetric results compared to the SRS NTO and ALDO optimization algorithms.

Predicting Prostate VMAT 3D Radiation Doses of Continuously Varying Organ Dose Trade-Offs Using a Conditional Variational Autoencoder

Predicting 3D radiation doses from planning structures is a promising method of knowledge-based treatment planning. However, most models are designed to predict only one 3D dose distribution per patient, based on historical organ dose trade-offs. To allow customizable plan generation, in this study, we aim to show the feasibility of dose prediction in which the degrees of organ dose trade-off could be explicitly specified. Specifically, the bladder vs. rectum dose trade-off in prostate cancer was investigated. In an IRB-approved study, we obtained imaging and structure contours for 167 patients with prostate cancer who received definitive radiotherapy. Training data was generated by automatically creating 3 different plans for each patient: while keeping target dose patterns constant, 1 base plan was generated with optimization objectives directly based on the output of a custom RapidPlan model prediction (S = 0), 1 plan with the goal to significantly lower bladder dose relative to the rectum (S = -1), and 1 plan with the goal to significantly lower rectum dose relative to the bladder (S = 1). This process was achieved by adjusting priority values during optimization. S is a scalar indicating the degree of bladder vs. rectum dose trade-off (higher S = higher dose to the bladder relative to the rectum). A conditional variational autoencoder (cVAE) was constructed as the generative model. Training, validation, and testing sets consist of 124, 10, and 33 patients, respectively. During training, the inputs to the model were 3D structure masks with voxel values modified based on S, and the output was the corresponding 3D dose. For model testing, we selected 7 equispaced values of S in the range [-1, +1] for each of the 33 test patients, generated the 3D doses for each S value (normalized to D2% = 110%), and calculated the differences of key dosimetric parameters (for S levels other than 0) compared to the predicted base plan (S = 0). The mean and standard deviations for these differences were reported. The cVAE model converged after training for 800 epochs. As the value of S increased from -1 to +1, the target coverage remained similar, while the doses to the bladder and rectum increased and decreased, respectively, as expected (Table 1). This pattern was also confirmed by qualitative examination of dose-volume histograms for additional S values. We demonstrated the feasibility of predicting 3D radiation dose distributions for prostate cancer where the degrees of organ dose trade-off could be explicitly specified.

A predicted three-dimensional dose sequence based treatment planning optimization method for gynecologic IMRT.

In knowledge-based treatment planning (KBTP) for intensity-modulated radiation therapy (IMRT), the quality of the plan is dependent on the sophistication of the predicted dosimetric information and its application. In this paper, we propose a KBTP method that based on the effective and reasonable utilization of a three-dimensional (3D) dose prediction on planning optimization. We used an organs-at-risk (OARs) dose distribution prediction model to create a voxel-based dose sequence based optimization objective for OARs doses. This objective was used to reformulate a traditional fluence map optimization model, which involves a tolerable spatial re-assignment of the predicted dose distribution to the OAR voxels based on their current doses' positions at a sorted dose sequencing. The feasibility of this method was evaluated with ten gynecology (GYN) cancer IMRT cases by comparing its generated plan quality with the original clinical plan. Results showed feasible plan by proposed method, with comparable planning target volume (PTV) dose coverage and greater dose sparing of the OARs. Among ten GYN cases, the average V30 and V45 of rectum were decreased by 4%±4% (p=0.02) and 4%±3% (p<0.01), respectively. V30 and V45 of bladder were decreased by 8%±2% (p<0.01) and 3%±2% (p<0.01), respectively. Our predicted dose sequence-based planning optimization method for GYN IMRT offered a flexible use of predicted 3D doses while ensuring the output plan consistency.

Can knowledge based treatment planning of VMAT for post-mastectomy locoregional radiotherapy involving internal mammary chain and supraclavicular fossa improve performance efficiency?

To validate and evaluate the performance of knowledge-based treatment planning for Volumetric Modulated Arc Radiotherapy for post-mastectomy loco-regional radiotherapy. Two knowledge-based planning (KBP) models for different dose prescriptions were built using the Eclipse RapidPlanTM v 16.1 (Varian Medical Systems, Palo Alto, USA) utilising the plans of previously treated patients with left-sided breast cancer who had undergone irradiation of the left chest wall, internal mammary nodal (IMN) region and supra-clavicular fossa (SCF). Plans of 60 and 73 patients were used to generate the KBP models for the prescriptions of 40 Gy in 15 fractions and 26 Gy in 5 fractions, respectively. A blinded review of all the clinical plans (CLI) and KBPs was done by two experienced radiation oncology consultants. Statistical analysis of the two groups was also done using the standard two-tailed paired t-test or Wilcoxon signed rank test, and p<0.05 was considered significant. A total of 20 metrics were compared. The KBPs were found to be either better (6/20) or comparable (10/20) to the CLIs for both the regimens. Dose to heart, contralateral breast,contralateral lung were either better or comparable in the KBP plans except of ipsilateral lung. Mean dose (Gy) for the ipsilateral lung are significantly (p˂0.001) higher in KBP though the values were acceptable clinically. Plans were of similar quality as per the result of the blinded review which was conducted by slice-by-slice evaluation of dose distribution for target coverage, overdose volume and dose to the OARs. However, it was also observed that treatment times in terms of monitoring units (MUs) and complexity indices are more in CLIs as compared with KBPs (p<0.001). KBP models for left-sided post-mastectomy loco-regional radiotherapy were developed and validated for clinical use. These models improved the efficiency of treatment delivery as well as work flow for VMAT planning involving both moderately hypo fractionated and ultra-hypo fractionated radiotherapy regimens.

Implementation, Dosimetric Assessment, and Treatment Validation of Knowledge-Based Planning (KBP) Models in VMAT Head and Neck Radiation Oncology

The aim of this study was to evaluate knowledge-based treatment planning (KBP) models in terms of their dosimetry and deliverability and to investigate their clinical benefits. Three H&N KBP models were built utilizing RapidPlan™, based on the dose prescription, which is given according to the planning target volume (PTV). The training set for each model consisted of 43 clinically acceptable volumetric modulated arc therapy (VMAT) plans. Model quality was assessed and compared to the delivered treatment plans using the homogeneity index (HI), conformity index (CI), structure dose difference (PTV, organ at risk-OAR), monitor units, MU factor, and complexity index. Model deliverability was assessed through a patient-specific quality assurance (PSQA) gamma index-based analysis. The dosimetric assessment showed better OAR sparing for the RapidPlan™ plans and for the low- and high-risk PTV, and the HI, and CI were comparable between the clinical and RapidPlan™ plans, while for the intermediate-risk PTV, CI was better for clinical plans. The 2D gamma passing rates for RapidPlan™ plans were similar or better than the clinical ones using the 3%/3 mm gamma-index criterion. Monitor units, the MU factors, and complexity indices were found to be comparable between RapidPlan™ and the clinical plans. Knowledge-based treatment plans can be safely adapted into clinical routines, providing improved plan quality in a time efficient way while minimizing user variability.

PO-1505: Knowledge based treatment planning and validation of VMAT for Cervical Cancer.

PO-1505: Knowledge based treatment planning and validation of VMAT for Cervical Cancer.

PO-1506: An automated knowledge based treatment planning solution for prostate VMAT

PO-1506: An automated knowledge based treatment planning solution for prostate VMAT

Influence of Cleaned-up Commercial Knowledge-Based Treatment Planning on Volumetric-Modulated Arc Therapy of Prostate Cancer.

Purpose:This study aimed to investigate the influence of cleaned-up knowledge-based treatment planning (KBP) models on the plan quality for volumetric-modulated arc therapy (VMAT) of prostate cancer.Materials and Methods:Thirty prostate cancer VMAT plans were enrolled and evaluated according to four KBP modeling methods as follows: (1) model not cleaned – trained by fifty other clinical plans (KBPORIG); (2) cases cleaned by removing plans that did not meet all clinical goals of the dosimetric parameters, derived from dose–volume histogram (DVH) (KBPC-DVH); (3) cases cleaned outside the range of ±1 standard deviation through the principal component analysis regression plots (KBPC-REG); and (4) cases cleaned using both methods (2) and (3) (KBPC-ALL). Rectal and bladder structures in the training models numbered 34 and 48 for KBPC-DVH, 37 and 33 for KBPC-REG, and 26 and 33 for KBPC-ALL, respectively. The dosimetric parameters for each model with one-time auto-optimization were compared.Results:All KBP models improved target dose coverage and conformity and provided comparable sparing of organs at risks (rectal and bladder walls). There were no significant differences in plan quality among the KBP models. Nevertheless, only the KBPC-ALL model generated no cases of >1% V78 Gy (prescribed dose) to the rectal wall, whereas the KBPORIG, KBPC-DVH, and KBPC-REG models included two, four, and three cases, respectively, which were difficult to overcome with KBP because the planning target volume (PTV) and rectum regions overlapped.Conclusions:The cleaned-up KBP model based on DVH and regression plots improved plan quality in the PTV–rectum overlap region.

Robustness comparative study of dose–volume–histogram prediction models for knowledge-based radiotherapy treatment planning

Purpose: To compare the robustness of three dose-volume-histogram (DVH) prediction models for knowledge-based treatment planning (KBP) for radiation therapy.Methods: Three models proposed by Zhu et...

OC-081: Knowledge based treatment planning and validation of VMAT for Cervical Cancer.

OC-081: Knowledge based treatment planning and validation of VMAT for Cervical Cancer.

Inter-planner variation in treatment-plan quality of plans created with a knowledge-based treatment planning system

PurposeThis study aimed to clarify the inter-planner variation of plan quality in knowledge-based plans created by nine planners. MethodsFive hypofractionated prostate-only (HPO) volumetric modulated arc therapy (VMAT) plans and five whole-pelvis (WP) VMAT plans were created by each planner using a knowledge-based planning (KBP) system. Nine planners were divided into three groups of three planners each: Senior, Junior, and Beginner. Single optimization with only priority modification for all objectives was performed to stay within the dose constraints. The coefficients of variation (CVs) for dosimetric parameters were evaluated, and a plan quality metric (PQM) was used to evaluate comprehensive plan quality. ResultsLower CVs (<0.05) were observed at dosimetric parameters in the planning target volume for both HPO and WP plans, while the CVs in the rectum and bladder for WP plans (<0.91) were greater than those for HPO plans (<0.17). The PQM values of HPO plans for Cases1–5 (average ± standard deviation) were 41.2 ± 7.1, 40.9 ± 5.6, and 39.9 ± 4.6 in the Senior, Junior, and Beginner groups, respectively. For the WP plans, the PQM values were 51.9 ± 6.3, 47.5 ± 4.3, and 40.0 ± 6.6, respectively. The number of clinically acceptable HPO and WP plans were 13/15 and 11/15 in the Senior group, 13/15 and 10/15 plans in the Junior group, and 8/15 and 2/15 plans in the Beginner group, respectively. ConclusionInter-planner variation in the plan quality with RapidPlan remains, especially for the complicated VMAT plans, due to planners’ heuristics.

Feasibility of Expediting Radiotherapy Clinical Workflow by Radiation Planning on Diagnostic Images

Studies have showed that delays from diagnosis to radiotherapy (RT) are detrimental to outcome for many cancers. The purpose of this study is to assess the feasibility of expediting clinical workflow by combining planning on diagnostic images and adaptive RT in order to eliminate treatment delays associated with RT simulation imaging. A cohort of 30 lung patients previously treated with RT were included. CT images acquired for diagnostic purposes on curved couch without proper immobilization for RT were used for diagnostic plans (Dd,d), employing knowledge-based treatment planning. The diagnostic plans were copied and recalculated on corresponding simulation CT images (Dd,s), which were the actual primary treatment datasets immobilized properly on flat therapy couch, to evaluate the likelihood of applying Dd,d for treatment directly. The diagnostic plans were scaled if needed to match PTV coverage at 95% of volume (Dd,s). Final plans (Ds,s) were reoptimized on the simulation CT images to best match the diagnostic plans, mocking an adaptive RT process. Changes in DVH were reported with Dd,d as the baseline. EUD-based tumor control probability (TCP) and normal tissue complication probability (NTCP) were reported to assess the consistency of clinical outcomes in all four scenarios. Ds,s matched Dd,d well in DVHs in all cases by visual check, while Dd,s and Dd,s matched Dd,d reasonably well in 21 out 30 cases. Scaling factor from Dd,s to Dd,s was 1.05 ± 0.04. Mean D95 in PTV changed slightly by -0.22% ± 0.41% (-0.12 ±0.24 Gy) from Dd,d to Ds,s, where mean changes of -0.14 ± 2.57 Gy, -0.92 ± 2.86 Gy, 0.99% ± 2.22%, 0.87 ± 1.3 Gy were observed in maximum cord dose, mean esophagus dose, lung V20 and mean heart dose, respectively. Mean TCP in Dd,d was 65.1% for PTV and changed by-0.02% from Dd,d to Ds,s. Mean NTCP in Dd,d were 0.02%, 5.67%, 0.32%, and 0.08%, and changed by 0.00%, 0.33%, 0.30%, and -0.02% for cord necrosis, esophageal perforation, pneumonitis and pericarditis, respectively, from Dd,d to Ds,s. Our results demonstrated that it is likely possible to expedite the clinical work flow by omitting simulation. It is feasible to plan on diagnostic CT images, then adapt during the treatment or treat directly or with a scaled plan depending on the treatment intent. This is regardless of patient positioning, couch types and immobilization devices.Abstract 1140; Table 1PTV D95 (%)Cord Max (Gy)Esophagus Mean (Gy)Lung V20 (%)Heart Mean (Gy)DVHDd,d100.3±1.737.9±8.321.1±8.321.1±7.28.2±5.4Dd,s92.5±8.438.8±10.120.4±7.524.2±16.19.7±5.9D’d,s96.8±6.140.7±10.721.4±8.124.9±16.110.2±6.4Ds,s100.1±1.537.8±8.120.2±7.322.1±7.99.1±5.7TCP/ NTCPPTVCord NecrosisEsophageal PerforationPneumonitisPericarditisDd,d65.1%±7.8%0.02%±0.03%5.7%±10.8%0.3%±0.7%0.08%±0.23%Dd,s47.3%±23.3%0.06%±0.15%4.4&%±9.2%4.0%±17.9%0.06%±0.175D’d,s50.5%±23.6%0.21%±0.55%10.8%±16.6%4.3%±18.2%0.15%±0.40%Ds,s65.1%±7.8%0.02%±0.02%6.0%±12.0%0.6%±1.6%0.06%±0.19% Open table in a new tab

Knowledge-based dose prediction models for head and neck cancer are strongly affected by interorgan dependency and dataset inconsistency.

The goal of this study was to generate a large treatment plan database for head and neck (H&N) cancer patients that can be considered as the gold standard to train and validate models for knowledge-based (KB) treatment planning and QA. With this dataset, the intrinsic prediction performance, the effect of interorgan dependency, and the impact of dataset inconsistency was investigated for an existing treatment planning QA model. The CT scans of 108 previously treated oropharyngeal patients were used to establish the plan database. For each patient, 15 Pareto optimal treatment plans with different planning priorities for the parotid glands were generated with fully automatic multicriterial treatment planning (1620 plans in total). For each of the 15 sets of plans in the database, a KB model was trained with 54 patients and validated on the other 54 by comparing the predictions with the achieved doses. The dose prediction accuracy (predicted-achieved) of the KB models was assessed and compared among the different models to characterize the intrinsic performance and effect of interorgan dependency. In addition, the effect of dataset inconsistency with respect to planning prioritizations was investigated by mixing plans with different prioritizations, for the training, the validation dataset, and for both combined. In the case of a high planning priority, the mean±SD of the prediction error for the mean dose of the parotid glands was only 0.2±2.2Gy, but this increased to 1.0±5.0Gy in the case that the parotid glands had a low planning priority. Dataset inconsistency (in planning priority) led to a large increase in prediction error for the parotid glands (mean±SD) from 0.2±2.2 Gy to 2.8±3.3Gy, -3.2±5.0Gy or -0.6±5.4Gy, depending on the way the datasets were mixed. The generated plan database can be used to validate and characterize KB prediction models for H&N cancer and will be made available upon request. The investigated KB model performed well in case the parotid glands had a high planning priority (little dependence on lower priority OARs), but poorly for organs for which the dose strongly depends on other higher priority OARs. To improve the performance of KB prediction models for H&N cancer, interorgan dependency should be modeled and accounted for. Dataset inconsistency has a large negative impact on the prediction errors of KB models and should be avoided as much as possible.

Mechanical performance of a commercial knowledge-based VMAT planning for prostate cancer

BackgroundThis study clarified the mechanical performance of volumetric modulated arc therapy (VMAT) plans for prostate cancer generated with a commercial knowledge-based treatment planning (KBP) and whether KBP system could be applied clinically without any major problems with mechanical performance.MethodsThirty consecutive prostate cancer patients who underwent VMAT using extant clinical plans were evaluated. The mechanical performance and dosimetric accuracy of the single optimized KBPs, which were trained with other 51 clinical plans, were compared with the clinical plans. The mechanical performance metrics were mean field area (MFA), mean asymmetry distance (MAD), cross-axis score (CAS), closed leaf score (CLS), small aperture score (SAS), leaf travel (LT), modulation complexity score (MCSv), and monitor unit (MU). The γ passing rates were evaluated with ArcCheck and EBT3 film.ResultsThe mean mechanical performance metrics (clinical plan vs. KBP) were as follows: 18.28 cm2 vs. 17.25 cm2 (MFA), 21.08 mm vs. 20.47 mm (MAD), 0.54 vs. 0.55 (CAS), 0.040 vs. 0.051 (CLS), 0.20 vs. 0.23 (SAS5mm), 458.5 mm vs. 418.8 mm (LT), 0.27 vs. 0.27 (MCSv), and 618.2 vs. 622.1 (MU), respectively. Significant differences were observed for CLS and LT. The average γ passing rates (clinical plan vs. KBP) were as follows: 99.0% vs. 99.1% (3%/3 mm) and 92.4% vs. 92.5% (2%/2 mm) with ArcCHeck, and 99.5% vs. 99.4% (3%/3 mm) and 95.2% vs. 95.4% (2%/2 mm) with EBT3 film, respectively.ConclusionsThe KBP used lower multileaf collimator (MLC) travel and more closed or small MLC apertures than the clinical plan. The KBP system of VMAT for the prostate cancer was acceptable for clinical use without any major problems.

Selection of external beam radiotherapy approaches for precise and accurate cancer treatment.

Physically precise external-beam radiotherapy (EBRT) technologies may not translate to the best outcome in individual patients. On the other hand, clinical considerations alone are often insufficient to guide the selection of a specific EBRT approach in patients. We examine the ways in which to compare different EBRT approaches based on physical, biological and clinical considerations, and how they can be enhanced with the addition of biophysical models and machine-learning strategies. The process of selecting an EBRT modality is expected to improve in tandem with knowledge-based treatment planning.

Knowledge-based treatment planning and its potential role in the transition between treatment planning systems

Commissioning a new treatment planning system (TPS) involves many time-consuming tasks. We investigated the role that knowledge-based planning (KBP) can play in aiding a clinic's transition to a new TPS. Sixty clinically treated prostate/prostate bed intensity-modulated radiation therapy (IMRT) plans were exported from an in-house TPS and were used to create a KBP model in a newly implemented commercial application. To determine the benefit that KBP may have in a TPS transition, the model was tested on 2 groups of patients. Group 1 consisted of the first 10 prostate/prostate bed patients treated in the commercial TPS after the transition from the in-house TPS. Group 2 consisted of 10 patients planned in the commercial TPS after 8 months of clinical use. The KBP-generated plan was compared with the clinically used plan in terms of plan quality (ability to meet planning objectives and overall dose metrics) and planning efficiency (time required to generate clinically acceptable plans). The KBP-generated plans provided a significantly improved target coverage (p = 0.01) compared with the clinically used plans for Group 1, but yielded plans of comparable target coverage to the clinically used plans for Group 2. For the organs at risk, the KBP-generated plans produced lower doses, on average, for every normal-tissue objective except for the maximum dose to 0.1 cc of rectum. The time needed for the KBP-generated plans ranged from 6 to 15 minutes compared to 30 to 150 and 15 to 60 minutes for manual planning in Groups 1 and 2, respectively. KBP is a promising tool to aid in the transition to a new TPS. Our study indicates that high-quality treatment plans could have been generated in the newly implemented TPS more efficiently compared with not using KBP. Even after 8 months of the clinical use, KBP still showed an increase in plan quality and planning efficiency compared with manual planning.

Leveraging an Expert Knowledge-Based Plan Matching System Enables Meeting More Clinical Objectives for Prostate External Beam Radiation Therapy

The creation of an ideal IMRT plan is a challenging and complex process utilizing a team of expert physicians, dosimetrists, medical physicists, and computer algorithms. Due to challenges with simultaneously optimizing multiple dosimetric indices, the full extent of achievable plan results (including plan trade-offs) are not known with high precision. Knowledge based treatment-planning (KBTP) aims to leverage past plan expertise to optimize new plans. The commercial software (QuickMatch™, Siris Medical Inc.) guides the dosimetrist through a library of previously approved plans to identify multiple achievable treatment plans for new patients. In general, the larger the database, the more insights that can be drawn from it. We report on treatment planning results from the largest known (to the best of our knowledge) plan database for treatment plan guidance of prostate plans. QuickMatch (QM) built a library with 1168 prostate patients (over 3,500 plans), spanning the years 2010-2016. These plans included pelvic node & non-pelvic node cases, with dose escalation to the Planning Target Volume (PTV) from 8100 to 8640 cGy using IMRT. We randomly identified 16 consecutive patients and planned them prospectively with QM. For each plan, a clinical dosimetrist using the same treatment planning system compared final dosimetry with respect to PTV coverages, bladder doses, rectal doses, and femur doses using standard planning techniques (SPT) compared to QM. We also retrospectively analyzed 1168 plans to identify those that did not meet either rectal, bladder, or femoral dose constraints using STP to determine if adding QM would result in improved dosimetry. Compared to STP alone, the addition of QM to STP improved overall dosimetry in the final prostate plans. In the 16 prospectively analyzed plans, nearly one more constraint per plan (mean difference = 0.8 constraints, matched pair t-test p=0.0044) was met without sacrificing other constraints or PTV coverage in every case (matched pair t-test p=0.86 that the means are different). QM guidance reduced rectal avoidance D10 in 85% of these patients. There were 561 instances where a clinical dose constraint was not met with STP. In these cases, QuickMatch improved the ability to meet clinical constraints, including PTV constraints, in 58% (323/561), in Rectal D10 in 67% (110/164), Rectal D50 in 59% (33/56), Bladder D10 in 54% (66/122), Bladder D50 in 25% (3/12), and femur D50 in 54% (111/207). We demonstrate the ability for an expert-trained knowledge based system to guide planning and to meet significantly more clinical dose constraints than standard planning for prostate RT.

Knowledge-based treatment planning: An inter-technique and inter-system feasibility study for prostate cancer

PurposeHelical Tomotherapy (HT) plans were used to create two RapidPlan knowledge-based (KB) models to generate plans with different techniques and to guide the optimization in a different treatment planning system for prostate plans. Feasibility and performance of these models were evaluated. Material and methodstwo sets of 35 low risk (LR) and 30 intermediate risk (IR) prostate cancer cases who underwent HT treatments were selected to train RapidPlan models. The KB predicted constraints were used to perform new 20KB plans using RapidArc technique (KB-RA) (inter-technique validation), and to optimise 20 new HT (KB-HT) plans in the Tomoplan (inter-system validation). For each validation modality, KB plans were benchmarked with the manual plans created by an expert planner (EP). ResultsRapidPlan was successfully configured using HT plans. The KB-RA plans fulfilled the clinical dose-volume requirements in 100% and 92% of cases for planning target volumes (PTVs) and organs at risk (OARs), respectively. For KB-HT plans these percentages were found to be a bit lower: 90% for PTVs and 86% for OARs. In comparison to EP plans, the KB-RA plans produced higher bladder doses for both LR and IR, and higher rectum doses for LR. KB-HT and EP plans produced similar results. ConclusionRapidPlan can be trained to create models by using plans of a different treatment modality. These models were suitable for generating clinically acceptable plans for inter-technique and inter-system applications. The use of KB models based on plans of consolidated technique could be useful with a new treatment modality.

The model training of knowledge-based radiotherapy treatment planning and its preliminary application

Objective Knowledge-based radiotherapy (KBRT) can reduce the plan quality variability induced by different experiences between physicians and improve the quality of treatment plans. Methods The Varian Rapid Plan system was used to train a dose-volume histogram (DVH) prediction model. The obtained model was preliminarily applied to semi-automatic design of the preoperative treatment plans for rectal cancer. Eighty high-quality volumetric modulated arc therapy plans were imported into the model training set of the Rapid Plan system. The structures of the plans were matched to the corresponding labels and codes as listed in the library. The training started after the verification of prescription. The residual plots, regression curves, geometric plots for organ at risk (OAR), in-field DVH plots, and model training logs were examined. After removal of the mismatch, the original plans were assessed to rule out outliers and influential data points. More similar plans may be added for another round of training. Ten KBRT plans were designed using the final model and compared with the clinical plans. Results For the two major OARs, the femoral head and bladder, the average goodness of fit of the principal component were 0.999415/1.0 and 0.999963/1.0 for the DVH model, and 0.999651/1.0 and 0.999945/1.0 for geometry-based expected dose model, respectively. In all the plans, 11 had Cook’s distance values exceeding the tolerance and 4 had studentized residual values exceeding the tolerance. The outliers were all kept in the training set to generalize the scope of the model. The 10 KBRT plans had significantly improved homogeneity indices for PGTV and PTV than the original plans (P=0.00, 0.04). The 10 KBRT plans also had significantly reduced D50% to the femoral head and bladder as well as significantly reduced mean doses to the bladder than the original plans (P=0.042, 0.000, 0.005). Conclusions In this study, the Rapid Plan system is used to train a KBRT model for design of preoperative radiotherapy plans for rectal cancer. The results of preliminary application meet the clinical requirements. Key words: Radiotherapy planning; Model training; Model verification