A Kalet

A state-based probabilistic model for tumor respiratory motion prediction

This work proposes a new probabilistic mathematical model for predicting tumor motion and position based on a finite state representation using the natural breathing states of exhale, inhale and end of exhale. Tumor motion was broken down into linear breathing states and sequences of states. Breathing state sequences and the observables representing those sequences were analyzed using a hidden Markov model (HMM) to predict the future sequences and new observables. Velocities and other parameters were clustered using a k-means clustering algorithm to associate each state with a set of observables such that a prediction of state also enables a prediction of tumor velocity. A time average model with predictions based on average past state lengths was also computed. State sequences which are known a priori to fit the data were fed into the HMM algorithm to set a theoretical limit of the predictive power of the model. The effectiveness of the presented probabilistic model has been evaluated for gated radiation therapy based on previously tracked tumor motion in four lung cancer patients. Positional prediction accuracy is compared with actual position in terms of the overall RMS errors. Various system delays, ranging from 33 to 1000 ms, were tested. Previous studies have shown duty cycles for latencies of 33 and 200 ms at around 90% and 80%, respectively, for linear, no prediction, Kalman filter and ANN methods as averaged over multiple patients. At 1000 ms, the previously reported duty cycles range from approximately 62% (ANN) down to 34% (no prediction). Average duty cycle for the HMM method was found to be 100% and 91 ± 3% for 33 and 200 ms latency and around 40% for 1000 ms latency in three out of four breathing motion traces. RMS errors were found to be lower than linear and no prediction methods at latencies of 1000 ms. The results show that for system latencies longer than 400 ms, the time average HMM prediction outperforms linear, no prediction, and the more general HMM-type predictive models. RMS errors for the time average model approach the theoretical limit of the HMM, and predicted state sequences are well correlated with sequences known to fit the data.

Validation of the pinnacle3 photon convolution-superposition algorithm applied to fast neutron beams

We evaluate a photon convolution‐superposition algorithm used to model a fast neutron therapy beam in a commercial treatment planning system (TPS). The neutron beam modeled was the Clinical Neutron Therapy System (CNTS) fast neutron beam produced by 50 MeV protons on a Be target at our facility, and we implemented the Pinnacle3 dose calculation model for computing neutron doses. Measured neutron data were acquired by an IC30 ion chamber flowing 5 cc/min of tissue equivalent gas. Output factors and profile scans for open and wedged fields were measured according to the Pinnacle physics reference guide recommendations for photon beams in a Wellhofer water tank scanning system. Following the construction of a neutron beam model, computed doses were then generated using 100 monitor units (MUs) beams incident on a water‐equivalent phantom for open and wedged square fields, as well as multileaf collimator (MLC)‐shaped irregular fields. We compared Pinnacle dose profiles, central axis doses, and off‐axis doses (in irregular fields) with 1) doses computed using the Prism treatment planning system, and 2) doses measured in a water phantom and having matching geometry to the computation setup. We found that the Pinnacle photon model may be used to model most of the important dosimetric features of the CNTS fast neutron beam. Pinnacle‐calculated dose points among open and wedged square fields exhibit dose differences within 3.9 cGy of both Prism and measured doses along the central axis, and within 5 cGy difference of measurement in the penumbra region. Pinnacle dose point calculations using irregular treatment type fields showed a dose difference up to 9 cGy from measured dose points, although most points of comparison were below 5 cGy. Comparisons of dose points that were chosen from cases planned in both Pinnacle and Prism show an average dose difference less than 0.6%, except in certain fields which incorporate both wedges and heavy blocking of the central axis. All clinical cases planned in both Prism and Pinnacle were found to be comparable in terms of dose‐volume histograms and spatial dose distribution following review by the treating clinicians. Variations were considered minor and within clinically acceptable limits by the treating clinicians. The Pinnacle TPS has sufficient computational modeling ability to adequately produce a viable neutron model for clinical use in treatment planning. PACS numbers: 87.53 Bn, 28.20.Pr, 87.53.Bn

SU-E-T-51: Bayesian Network Models for Radiotherapy Error Detection

PURPOSE To develop a probabilistic model of radiotherapy plans using Bayesian networks that will detect potential errors in radiation delivery. METHODS Semi-structured interviews with medical physicists and other domain experts were employed to generate a set of layered nodes and arcs forming a Bayesian Network (BN) which encapsulates relevant radiotherapy concepts and their associated interdependencies. Concepts in the final network were limited to those whose parameters are represented in the institutional database at a level significant enough to develop mathematical distributions. The concept-relation knowledge base was constructed using the Web Ontology Language (OWL) and translated into Hugin Expert Bayes Network files via the the RHugin package in the R statistical programming language. A subset of de-identified data derived from a Mosaiq relational database representing 1937 unique prescription cases was processed and pre-screened for errors and then used by the Hugin implementation of the Estimation-Maximization (EM) algorithm for machine learning all parameter distributions. Individual networks were generated for each of several commonly treated anatomic regions identified by ICD-9 neoplasm categories including lung, brain, lymphoma, and female breast. RESULTS The resulting Bayesian networks represent a large part of the probabilistic knowledge inherent in treatment planning. By populating the networks entirely with data captured from a clinical oncology information management system over the course of several years of normal practice, we were able to create accurate probability tables with no additional time spent by experts or clinicians. These probabilistic descriptions of the treatment planning allow one to check if a treatment plan is within the normal scope of practice, given some initial set of clinical evidence and thereby detect for potential outliers to be flagged for further investigation. CONCLUSION The networks developed here support the use of probabilistic models into clinical chart checking for improved detection of potential errors in RT plans.

Dosimetric effects of bolus and lens shielding in treating ocular lymphomas with low-energy electrons

Radiation therapy is an effective treatment for primary orbital lymphomas. Lens shielding with electrons can reduce the risk of high-grade cataracts in patients undergoing treatment for superficial tumors. This work evaluates the dosimetric effects of a suspended eye shield, placement of bolus, and varying electron energies. Film (GafChromic EBT3) dosimetry and relative output factors were measured for 6, 8, and 10 MeV electron energies. A customized 5-cm diameter circle electron orbital cutout was constructed for a 6 × 6-cm applicator with a suspended lens shield (8-mm diameter Cerrobend cylinder, 2.2-cm length). Point doses were measured using a scanning electron diode in a solid water phantom at depths representative of the anterior and posterior lens. Depth dose profiles were compared for 0-mm, 3-mm, and 5-mm bolus thicknesses. At 5 mm (the approximate distance of the anterior lens from the surface of the cornea), the percent depth dose under the suspended lens shield was reduced to 15%, 15%, and 14% for electron energies 6, 8, and 10 MeV, respectively. Applying bolus reduced the benefit of lens shielding by increasing the estimated doses under the block to 27% for 3-mm and 44% for 5-mm bolus for a 6 MeV incident electron beam. This effect is minimized with 8 MeV electron beams where the corresponding values were 15.5% and 18% for 3-mm and 5-mm bolus. Introduction of a 7-mm hole in 5-mm bolus to stabilize eye motion during treatment altered lens doses by about 1%. Careful selection of electron energy and consideration of bolus effects are needed to account for electron scatter under a lens shield.

Developing Bayesian networks from a dependency‐layered ontology: A proof‐of‐concept in radiation oncology

Purpose: Bayesian networks (BNs) are graphical representations of probabilistic knowledge that offer normative reasoning under uncertainty and are well suited for use in medical domains. Traditional knowledge‐based network development of BN topology requires that modeling experts establish relevant dependency links between domain concepts by searching and translating published literature, querying domain experts, or applying machine learning algorithms on data. For initial development these methods are time‐intensive and this cost hinders the growth of BN applications in medical decision making. Further, this approach fails to utilize knowledge representation in medical fields to automate network development. Our research alleviates the challenges surrounding BN modeling in radiation oncology by leveraging an ontology based hub and spoke system for BN construction. Methods: We implement a hub and spoke system by developing (a) an ontology of knowledge in radiation oncology (the hub) which includes dependency semantics similar to BN relations and (b) a software tool that operates on ontological semantics using deductive reasoning to create BN topologies (the spokes). We demonstrate that network topologies built using the software are terminologically consistent and form networks that are topologically compatible with existing ones. We do this first by merging two different BN models for prostate cancer radiotherapy prediction which contain domain cross terms. We then use the logic to perform discovery of new causal chains between radiation oncology concepts. Results: From the radiation oncology (RO) ontology we successfully reconstructed a previously published prostate cancer radiotherapy Bayes net using up‐to‐date domain knowledge. Merging this model with another similar prostate cancer model in the RO domain produced a larger, highly interconnected model representing the expanded scope of knowledge available regarding prostate cancer therapy parameters, complications, and outcomes. The causal discovery resulted in an automatically‐built causal network model of all ontologized radiotherapy concepts between a ‘Mucositis’ complication and anatomic tumor location. Conclusions: The proposed model building approach lowers barriers to developing probabilistic models relevant to real‐world clinical decision making, and offers a solution to the consistency and compatibility problems. Further, the knowledge representation in this work demonstrates potential for broader radiation oncology applications outside of Bayes nets.

SU-G-TeP1-11: Predictors of Cardiac and Lung Dose Sparing in DIBH for Left Breast Treatment

PURPOSE This retrospective study of left sided whole breast radiation therapy (RT) patients investigates possible predictive parameters correlating to cardiac and left lung dose sparing by deep inspiration breath-hold (DIBH) technique compared to free-breathing (FB). METHODS Thirty-one patients having both DIBH and FB CT scans were included in the study. All patients were planned with a standard step-and-shoot tangential technique using MV photons, with prescription of 50Gy or 50.4Gy. The displacement of the breath hold sternal mark during DIBH, the cardiac contact distances of the axial (CCDax) and parasagittal (CCDps) planes, and lateral-heart-to-chest (LHC) distance on FB CT scans were measured. Lung volumes, mean dose and dose-volume histograms (V5, V10 and V20) were analyzed and compared for heart and left lung for both FB and DIBH techniques. Correlation analysis was performed to identify the predictors for heart and left lung dose sparing. Two-tailed Students t-test and linear regression were used for data analysis with significance level of P≤0.05. RESULTS All dosimetric metrics for the heart and left lung were significantly reduced (P<0.01) with DIBH. Breath hold sternal mark displacement ranged from 0.4-1.8 cm and correlated with mean (P=0.05) and V5 (P=0.02) of heart dose reduction by DIBH. FB lung volume showed correlation with mean lung dose reduction by DIBH (P<0.01). The FB-CCDps and FB-LHC distance had strong positive and negative correlation with FB mean heart dose (P<0.01) and mean heart dose reduction by DIBH (P<0.01), respectively. FB-CCDax showed no correlation with dosimetric changes. CONCLUSION DIBH technique has been shown to reduce dose to the heart and left lung. In this patient cohort, FB-CCDps, FB-LHC distance, and FB lung volume served as significant predictors for heart and left lung. These parameters can be further investigated to be used as a tool to better select patients who will benefit from DIBH.

WE-H-BRC-06: A Unified Machine-Learning Based Probabilistic Model for Automated Anomaly Detection in the Treatment Plan Data

PURPOSE The purpose of this work was to investigate the ability of a machine-learning based probabilistic approach to detect radiotherapy treatment plan anomalies given initial disease classes information. METHODS In total we obtained 1112 unique treatment plans with five plan parameters and disease information from a Mosaiq treatment management system database for use in the study. The plan parameters include prescription dose, fractions, fields, modality and techniques. The disease information includes disease site, and T, M and N disease stages. A Bayesian network method was employed to model the probabilistic relationships between tumor disease information, plan parameters and an anomaly flag. A Bayesian learning method with Dirichlet prior was useed to learn the joint probabilities between dependent variables in error-free plan data and data with artificially induced anomalies. In the study, we randomly sampled data with anomaly in a specified anomaly space.We tested the approach with three groups of plan anomalies - improper concurrence of values of all five plan parameters and values of any two out of five parameters, and all single plan parameter value anomalies. Totally, 16 types of plan anomalies were covered by the study. For each type, we trained an individual Bayesian network. RESULTS We found that the true positive rate (recall) and positive predictive value (precision) to detect concurrence anomalies of five plan parameters in new patient cases were 94.45±0.26% and 93.76±0.39% respectively. To detect other 15 types of plan anomalies, the average recall and precision were 93.61±2.57% and 93.78±3.54% respectively. The computation time to detect the plan anomaly of each type in a new plan is ∼0.08 seconds. CONCLUSION The proposed method for treatment plan anomaly detection was found effective in the initial tests. The results suggest that this type of models could be applied to develop plan anomaly detection tools to assist manual and automated plan checks. The senior author received research grants from ViewRay Inc. and Varian Medical System.

Electron beam energy QA — a note on measurement tolerances

Monthly QA is recommended to verify the constancy of high‐energy electron beams generated for clinical use by linear accelerators. The tolerances are defined as 2%/2 mm in beam penetration according to AAPM task group report 142. The practical implementation is typically achieved by measuring the ratio of readings at two different depths, preferably near the depth of maximum dose and at the depth corresponding to half the dose maximum. Based on beam commissioning data, we show that the relationship between the ranges of energy ratios for different electron energies is highly nonlinear. We provide a formalism that translates measurement deviations in the reference ratios into change in beam penetration for electron energies for six Elekta (6‐18 MeV) and eight Varian (6‐22 MeV) electron beams. Experimental checks were conducted for each Elekta energy to compare calculated values with measurements, and it was shown that they are in agreement. For example, for a 6 MeV beam a deviation in the measured ionization ratio of ±15% might still be acceptable (i.e., be within ±2 mm), whereas for an 18 MeV beam the corresponding tolerance might be ±6%. These values strongly depend on the initial ratio chosen. In summary, the relationship between differences of the ionization ratio and the corresponding beam energy are derived. The findings can be translated into acceptable tolerance values for monthly QA of electron beam energies. PACS number(s): 87.55, 87.56Monthly QA is recommended to verify the constancy of high-energy electron beams generated for clinical use by linear accelerators. The tolerances are defined as 2%/2 mm in beam penetration according to AAPM task group report 142. The practical implementation is typically achieved by measuring the ratio of readings at two different depths, preferably near the depth of maximum dose and at the depth corresponding to half the dose maximum. Based on beam commissioning data, we show that the relationship between the ranges of energy ratios for different electron energies is highly nonlinear. We provide a formalism that translates measurement deviations in the reference ratios into change in beam penetration for electron energies for six Elekta (6-18 MeV) and eight Varian (6-22 MeV) electron beams. Experimental checks were conducted for each Elekta energy to compare calculated values with measurements, and it was shown that they are in agreement. For example, for a 6 MeV beam a deviation in the measured ionization ratio of ±15% might still be acceptable (i.e., be within ±2 mm), whereas for an 18 MeV beam the corresponding tolerance might be ±6%. These values strongly depend on the initial ratio chosen. In summary, the relationship between differences of the ionization ratio and the corresponding beam energy are derived. The findings can be translated into acceptable tolerance values for monthly QA of electron beam energies. PACS number(s): 87.55, 87.56.

SU-E-T-239: In Vivo Dosimetry with Surface Diodes during Total Body Irradiation: A Patient Thickness Factor to Correct Midline Dose

PURPOSE In vivo dosimetry (IVD) assessment of treatment dose is important when delivering total body irradiation (TBI). One method is to average AP and PA surface diode measurements and compare them to prescribed midline doses. We designed phantom studies to examine the impact of patient thickness on surface IVD measurements under TBI conditions. METHODS Phantom studies were designed to assess the effects of patient thickness on diode IVD. Sun Nuclear QED diodes with inherent buildup were placed on anterior and posterior surfaces of a solid water phantom. Phantom thickness was varied between 20 and 40 cm. A PTW farmer chamber was inserted in the center of the phantom at 425 SSD to reflect prescribed midline dose, and 50 cGy was delivered to midline with 18 MV photons. Averaged entrance and exit diode doses were then compared to farmer chamber measurements of phantom midline dose. RESULTS A trend of increased deviation with increasing umbilicus thickness was observed between averaged surface diodes and midline farmer chamber measurements. Averaged surface diode dose ranged from 49.6 cGy (20 cm thickness) to 52.1 cGy (40 cm thickness). Interpolation of diode measurements to midline resulted in linear overestimation of delivered dose relative to farmer chamber measurements at midline, up to 6.8% at 40 cm umbilicus thickness. CONCLUSION Accurate in vivo dosimetry at time of patient TBI is important to allow individual correction of MU exposure and tissue compensation. Without patient thickness correction, overresponse of surface diodes may lead to unnecessary clinical intervention to treatment MU or compensation and insufficient midline dose. Additionally, SAD setup is preferable to SSD setup to minimize thickness non-linearity. In conclusion, thickness correction factors should be used to generate expected diode readings for patients with thickness greater than 30 cm.

SU‐E‐T‐718: Modeling a Fast Neutron Therapy Beam with a Convolution/superposition Algorithm

Purpose: To determine if a photon convolution/superposition algorithm could be used to model a fast neutron therapy beam in a commercial treatment planning system. Methods: The beam to be modeled was the Clinical Neutron Therapy System (CNTS) fast neutron beam produced by 50 MeV protons on a Be target at the University of Washington(UW) Medical Center. The dose calculation model was that implemented in Pinnacle, 3 (Philips Medical Systems). Measured neutron dose data were acquired with an IC30 ion chamber with tissue equivalent gas. The neutron beam was modeled using the auto‐modeling tools available in the Pinnacle system for photon beams. Doses were then computed using a 100 MU beam incident on a water equivalent phantom for open and wedged square fields as well as MLC shaped irregular fields. Pinnacle generated profiles and central axis dose points were compared to two sets of doses: 1) doses computed with the UW PRISMtreatment planning system in the same geometry as the Pinnacle setup and 2) doses measured in a water tank. Results: The Pinnacle photonmodel incorporates most of the important dosimetric features of the neutron beam. Computed doses compared well to both the Prism TPS and measured data. We found that calculated dose points among open and wedged square fields were within 2% of both Prism and measured doses along the central axis, and within 5% of measurement in the penumbra region. Dose point calculations using irregular treatment type fields were within 3% of measured dose points. Conclusion: The Pinnacle TPS has sufficient computational modeling ability to adequately produce a viable neutronmodel for potential use in clinical treatment planning. Further testing of irregular fields must be performed and results analyzed prior to implementation in the clinic.