Qingya Zhao

Gating based on internal/external signals with dynamic correlation updates.

Precise localization of mobile tumor positions in real time is critical to the success of gated radiotherapy. Tumor positions are usually derived from either internal or external surrogates. Fluoroscopic gating based on internal surrogates, such as implanted fiducial markers, is accurate however requiring a large amount of imaging dose. Gating based on external surrogates, such as patient abdominal surface motion, is non-invasive however less accurate due to the uncertainty in the correlation between tumor location and external surrogates. To address these complications, we propose to investigate an approach based on hybrid gating with dynamic internal/external correlation updates. In this approach, the external signal is acquired at high frequency (such as 30 Hz) while the internal signal is sparsely acquired (such as 0.5 Hz or less). The internal signal is used to validate and update the internal/external correlation during treatment. Tumor positions are derived from the external signal based on the newly updated correlation. Two dynamic correlation updating algorithms are introduced. One is based on the motion amplitude and the other is based on the motion phase. Nine patients with synchronized internal/external motion signals are simulated retrospectively to evaluate the effectiveness of hybrid gating. The influences of different clinical conditions on hybrid gating, such as the size of gating windows, the optimal timing for internal signal acquisition and the acquisition frequency are investigated. The results demonstrate that dynamically updating the internal/external correlation in or around the gating window will reduce false positive with relatively diminished treatment efficiency. This improvement will benefit patients with mobile tumors, especially greater for early stage lung cancers, for which the tumors are less attached or freely floating in the lung.

Statistical analysis and correlation discovery of tumor respiratory motion

Tumors, especially in the thorax and abdomen, are subject to respiratory motion, and understanding the structure of respiratory motion is a key component to the management and control of disease in these sites. We have applied statistical analysis and correlation discovery methods to analyze and mine tumor respiratory motion based on a finite state model of tumor motion. Aggregates (such as minimum, maximum, average and mean), histograms, percentages, linear regression and multi-round statistical analysis have been explored. The results have been represented in various formats, including tables, graphs and text description. Different graphs, for example scatter plots, clustered column figures, 100% stacked column figures and box-whisker plots, have been applied to highlight different aspects of the results. The internal tumor motion from 42 lung tumors, 30 of which have motion larger than 5 mm, has been analyzed. Results for both inter-patient and intra-patient motion characteristics, such as duration and travel distance patterns, are reported. New knowledge of patient-specific tumor motion characteristics have been discovered, such as expected correlations between properties. The discovered tumor motion characteristics will be utilized in different aspects of image-guided radiation treatment, including treatment planning, online tumor motion prediction and real-time radiation dose delivery.

A sector-integration method for dose/MU calculation in a uniform scanning proton beam.

An accurate, simple and time-saving sector integration method for calculating the proton output (dose/monitor unit, MU) is presented based on the following treatment field parameters: aperture shape, aperture size, measuring position, beam range and beam modulation. The model is validated with dose/MU values for 431 fields previously measured at our center. The measurements were obtained in a uniform scanning proton beam with a parallel plate ionization chamber in a water phantom. For beam penetration depths of clinical interest (6-27 cm water), dose/MU values were measured as a function of spread-out Bragg peak (SOBP) extent and aperture diameter. First, 90 randomly selected fields were used to derive the model parameters, which were used to compute the dose/MU values for the remaining 341 fields. The min, max, average and the standard deviation of the difference between the calculated and the measured dose/MU values of the 341 fields were used to evaluate the accuracy and stability, for different energy ranges, aperture sizes, measurement positions and SOBP values. The experimental results of the five different functional sets showed that the calculation model is accurate with calculation errors ranging from -2.4% to 3.3%, and 99% of the errors are less than +/-2%. The accuracy increases with higher energy, larger SOBP and bigger aperture size. The average error in the dose/MU calculation for small fields (field size <25 cm(2)) is 0.31 +/- 0.96 (%).

Proton Therapy Facility Planning From a Clinical and Operational Model.

This paper provides a model for planning a new proton therapy center based on clinical data, referral pattern, beam utilization and technical considerations. The patient-specific data for the depth of targets from skin in each beam angle were reviewed at our center providing megavoltage photon external beam and proton beam therapy respectively. Further, data on insurance providers, disease sites, treatment depths, snout size and the beam angle utilization from the patients treated at our proton facility were collected and analyzed for their utilization and their impact on the facility cost. The most common disease sites treated at our center are head and neck, brain, sarcoma and pediatric malignancies. From this analysis, it is shown that the tumor depth from skin surface has a bimodal distribution (peak at 12 and 26 cm) that has significant impact on the maximum proton energy, requiring the energy in the range of 130-230 MeV. The choice of beam angles also showed a distinct pattern: mainly at 90° and 270°; this indicates that the number of gantries may be minimized. Snout usage data showed that 70% of the patients are treated with 10 cm snouts. The cost of proton beam therapy depends largely on the type of machine, maximum beam energy and the choice of gantry versus fixed beam line. Our study indicates that for a 4-room center, only two gantry rooms could be needed at the present pattern of the patient cohorts, thus significantly reducing the initial capital cost. In the USA, 95% and 100% of patients can be treated with 200 and 230 MeV proton beam respectively. Use of multi-leaf collimators and pencil beam scanning may further reduce the operational cost of the facility.

Quality Assurance of Proton Compensators

Distal dose conformity to a target volume is achieved by choosing proper range of the proton beam and modifying it by custom designed compensator. The voxel of the compensator is milled precisely using a computerized milling machine. The quality assurance of proton compensators used in patient treatment is a vital issue in proton therapy that is investigated. Forty one patient specific proton compensators were analyzed that were manufactured using Milltronics VM16 computerized machine. Comparisons were made between the treatment planning system output data file and the voxel depth of the compensator. Point depths were measured by Sheffield Discovery II and were input into OmniPro IMRT software. The spatial distributions and histograms of compensator depth differences were visualized and evaluated. The acceptable depth difference between the measured and specified depth 0.32mm was maintained. The acceptance criterion of the manufactured compensator was 95% data points. For 37 of the 41 compensators, all measured points passed the test. The average difference is ±0.04 mm with the max difference depth of 0.28mm. One compensator was rejected as 10.4% of the measured points failed the test. For 3 other compensators, about 2.7% of the measured points failed the test with the maximum difference of 9.8 mm. Spatial distribution of the poor quality points was evaluated to decide if the compensa- tors should be rejected. A method for proton compensator quality assurance based on an existing commercial software system is introduced. The QA approach is coupled with visualization and statistical evaluation. Spatial distributions of error points can be identified, which are valuable for further dose distribution analysis.

Knowledge Discovery from Tumor Respiratory Motion Data

Image-guided radiation treatment (IGRT) is a recent advancement in the treatment of cancer patients with tumors in the abdomen or lungs. However, the efficacy of radiation treatment in these locations is often degraded by tumor respiratory motion. Therefore, the characterization and prediction of tumor motion are critical for precise cancer radiation treatment. This paper describes an approach for knowledge discovery from respiratory motion according to different motion properties. A hierarchical data model is proposed for tumor motion data representation. Various statistical analysis and correlation discovery over complex tumor respiratory motion data are designed based on a data cube to characterize different tumor motion properties. The outcomes will provide quantitative information for tumor motion prediction and real-time treatment delivery, which results better care for cancer patients.

A line-profile based double partial fusion method for acquiring planning CT of oversized patients in radiation treatment

True 3D CT dataset for treatment planning of an oversized patient is difficult to acquire due to the bore size and field of view (FOV) reconstruction. This project aims to provide a simple approach to reconstruct true CT data for oversize patients using CT scanner with limited FOV by acquiring double partial CT (left and right side) images. An efficient line profile‐based method has been developed to minimize the difference of the CT numbers in the overlapping region between the right and left images and to generate a complete true 3D CT dataset in the natural state. New image processing modules have been developed and integrated to the Insight Segmentation & Registration Toolkit (ITK 3.6) package. For example, different modules for image cropping, line profile generation, line profile matching, and optimized partial image fusion have been developed. The algorithm has been implemented for images containing the bony structure of the spine and tested on 3D CT planning datasets from both phantom and real patients with satisfactory results in both cases. The proposed optimized line profile‐based partial registration method provides a simple and accurate method for acquiring a complete true 3D CT dataset for an oversized patient using CT scanning with small bore size, that can be used for accurate treatment planning. PACS number: 89

Dose monitoring and output correction for the effects of scanning field changes with uniform scanning proton beam

PURPOSE The output of a proton beam is affected by proton energy, Spread-Out Bragg Peak (SOBP) width, aperture size, dose rate, and the point of measurement. In a uniform scanning proton beam (USPB), the scanning field size is adjusted (including the vertical length and the horizontal width) according to the treatment field size with appropriate margins to reduce secondary neutron production. Different scanning field settings result in beam output variations that are investigated in this study. METHODS The measurements are performed with a parallel plate Markus chamber at the center of SOBP under the reference condition with 16 cm range, 10 cm SOBP, and 5 cm air gap. The effect of dose rate on field output is studied by varying proton beam current from 0.5 to 7 nA. The effects of scanning field settings are studied by varying independently the field width and length from 12 x 12 to 30 x 30 cm2. RESULTS The results demonstrate that scanning field variations can produce output variation up to 3.80%. In addition, larger output variation is observed with scanning field changes along the stem direction of the patient dose monitor (PDM). By investigating the underlying physics of incomplete charge collection and the stem effects of the PDM, an analytical model is proposed to calculate USPB output with consideration of the scanning field area and the PDM stem length that is irradiated. The average absolute difference between the measured output and calculated output using our new correction model are within 0.13 and 0.08% for the 20 and 30 cm snouts, respectively. CONCLUSIONS This study proposes a correction model for accurate USPB output calculation, which takes account of scanning field settings and the PDM stem effects. This model may be used to extend the existing output calculation model from one snout size to other snout sizes with customized scanning field settings. The study is especially useful for calculating field output for treatment without individualized patient specific measurements.

Subsequence based treatment failure detection and intervention in image guided radiotherapy

Respiratory motion induces discrepancy between the expected tumor positions used in treatment planning and the actual positions during treatment delivery. Such motion degrades greatly the effectiveness of the radiation treatment. To address this challenge, we have proposed an online treatment failure detection approach with image guidance. Tumor motion is tracked in real-time during treatment delivery and compared to the baseline motion used in treatment planning. Tracking errors are recovered online with subdivided subsequence correlation. A stop-n-wait dose delivery procedure is applied to minimize treatment errors. Two approaches have been developed to address baseline shift in tumor motion. The performances are evaluated using three different metrics: the misplacement of the tumor, the treatment efficacy, and the intervention frequency. The results showed that the new approaches will reduce treatment errors, improve dose delivery efficiency, and reduce treatment interventions. This study has the potential to be employed in clinical practice thus improving radiation outcome.

SU‐C‐BRCD‐05: A Failure Mode and Effects Analysis (FMEA) Approach for Craniospinal Irradiation (CSI) with Proton Therapy (PT)

Purpose: In this paper, we report an FMEA approach on CSI at the IU Health Proton Therapy Center. Methods: A process map in proton CSI is developed. Each process consists of a number of sub‐processes. For each sub‐process, possible failures that may affect the process are identified and their respective Risk Priority Number (RPN) are calculated based on their severities (S), likelihood of occurrence (O) and probability of being detected (D), following the TG100 guidelines. Failures that most adversely impact the treatment are identified and quality assurance procedures to safeguard these most serious failures are developed. Attention is also given to certain failures which have low RPN but which may have dire consequence if it occurs in a treatment. Results: Ten intermediate processes involved in the CSI are identified. The number of sub‐processes within each intermediate process varies, from as few as one to as many as 11. For the ten processes in PTCSI, there are total of 66 sub‐processes, 139 failure modes and 561 causes of failures. 6/10 processes have failures with RPN=300. Examples of failures with such large RPN values are errors in administering anesthesia, in‐correct patient setup for image guidance, problems in the handling of apertures and compensators, etc. Conclusions: The implementation of FMEA in CSI (or any treatments) is a team effort. Significant efforts are involved in setting up process trees, failure modes, estimating the RPN values for each cause of failures, etc. However, once a FMEA is properly aligned, it is relatively easy to identify the most critical factors that require special attentions or QA to ensure safe execution of the processes. A learning curve to implement FMEA in any radiation oncology department should be expected given the different analysis practice from traditional QA approaches.