Practice patterns of palliative radiation therapy in pediatric oncology patients in an international pediatric research consortium.

BACKGROUND/OBJECTIVES: The practice of palliative radiation therapy (RT) is based on extrapolation from adult literature. We evaluated patterns of pediatric palliative RT to describe regimens used to identify opportunity for future pediatric-specific clinical trials. DESIGN/METHODS: Six international institutions with pediatric expertise completed a 122-item survey evaluating patterns of palliative RT for patients ≤21 years old from 2010 to 2015. Two institutions use proton RT. Palliative RT was defined as treatment with the goal of symptom control or prevention of immediate life-threatening progression. RESULTS: Of 3,225 pediatric patients, 365 (11%) were treated with palliative intent to a total of 427 disease sites. Anesthesia was required in 10% of patients. Treatment was delivered to metastatic disease in 54% of patients. Histologies included neuroblastoma (30%), osteosarcoma (18%), leukemia/lymphoma (12%), rhabdomyosarcoma (12%), medulloblastoma/ependymoma (12%), Ewing sarcoma (8%), and other (8%). Indications included pain (43%), intracranial symptoms (23%), respiratory compromise (14%), cord compression (8%), and abdominal distention (6%). Sites included nonspine bone (35%), brain (16% primary tumors, 6% metastases), abdomen/pelvis (15%), spine (12%), head/neck (9%), and lung/mediastinum (5%). Re-irradiation comprised 16% of cases. Techniques employed three-dimensional conformal RT (41%), intensity-modulated RT (23%), conventional RT (26%), stereotactic body RT (6%), protons (1%), electrons (1%), and other (2%). The most common physician-reported barrier to consideration of palliative RT was the concern about treatment toxicity (83%). CONCLUSION: There is significant diversity of practice in pediatric palliative RT. Combined with ongoing research characterizing treatment response and toxicity, these data will inform the design of forthcoming clinical trials to establish effective regimens and minimize treatment toxicity for this patient population.

A patient safety education program in a medical physics residency.

Education in patient safety and quality of care is a requirement for radiation oncology residency programs according to accrediting agencies. However, recent surveys indicate that most programs lack a formal program to support this learning. The aim of this report was to address this gap and share experiences with a structured educational program on quality and safety designed specifically for medical physics therapy residencies. Five key topic areas were identified, drawn from published recommendations on safety and quality. A didactic component was developed, which includes an extensive reading list supported by a series of lectures. This was coupled with practice-based learning which includes one project, for example, failure modes and effect analysis exercise, and also continued participation in the departmental incident learning system including a root-cause analysis exercise. Performance was evaluated through quizzes, presentations, and reports. Over the period of 2014-2016, five medical physics residents successfully completed the program. Evaluations indicated that the residents had a positive experience. In addition to educating physics residents this program may be adapted for medical physics graduate programs or certificate programs, radiation oncology residencies, or as a self-directed educational project for practicing physicists. Future directions might include a system that coordinates between medical training centers such as a resident exchange program.

Medical Physics Practice Guideline 4.a: Development, implementation, use and maintenance of safety checklists.

The American Association of Physicists in Medicine (AAPM) is a nonprofit professional society whose primary purposes are to advance the science, education and professional practice of medical physics. The AAPM has more than 8,000 members and is the principal organization of medical physicists in the United States.The AAPM will periodically define new practice guidelines for medical physics practice to help advance the science of medical physics and to improve the quality of service to patients throughout the United States. Existing medical physics practice guidelines will be reviewed for the purpose of revision or renewal, as appropriate, on their fifth anniversary or sooner.Each medical physics practice guideline represents a policy statement by the AAPM, has undergone a thorough consensus process in which it has been subjected to extensive review, and requires the approval of the Professional Council. The medical physics practice guidelines recognize that the safe and effective use of diagnostic and therapeutic radiology requires specific training, skills, and techniques, as described in each document. Reproduction or modification of the published practice guidelines and technical standards by those entities not providing these services is not authorized.The following terms are used in the AAPM practice guidelines:Must and Must Not: Used to indicate that adherence to the recommendation is considered necessary to conform to this practice guideline.Should and Should Not: Used to indicate a prudent practice to which exceptions may occasionally be made in appropriate circumstances.

Improved accuracy for noncoplanar radiotherapy: an EPID-based method for submillimeter alignment of linear accelerator table rotation with MV isocenter.

Accurate alignment of linear accelerator table rotational axis with radiation isocenter is critical for noncoplanar radiotherapy applications. The purpose of the present study is to develop a method to align the table rotation axis and the MV isocenter to submillimeter accuracy. We developed a computerized method using electronic portal imaging device (EPID) and measured alignment stability over time. Mechanical and radiation isocenter coincidence was measured by placing a steel ball bearing at radiation isocenter using existing EPID techniques. Then, EPID images were acquired over the range of table rotation. A MATLAB script was developed to calculate the center of rotation, as well as the necessary adjustment to move the table rotational axis to MV isocenter. Adjustment was applied via torque to screws at the base of the linac table. Stability of rotational alignment was measured with 49 measurements over 363 days on four linacs. Initial rotational misalignment from radiation isocenter ranged from 0.91-2.11 mm on the four tested linacs. Linac-A had greatest error (> 2 mm) and was adjusted with the described method. After adjustment, the error was significantly decreased to 0.40 ± 0.12 mm. The adjustment was stable over the course of 15 measurements over 231 days. Linac-B was not adjusted, but tracked from time of commissioning with 27 measurements over 363 days. No discernible shift in couch characteristics was observed (mean error 1.40 ± 0.22 mm). The greater variability for Linac-B may relate to the interchangeable two-piece couch, which allows more lateral movement than the one-piece Linac-A couch. Submillimeter isocenter alignment was achieved by applying a precision correction to the linac table base. Table rotational characteristics were shown to be stable over the course of twelve months. The accuracy and efficiency of this method may make it suitable for acceptance testing, annual quality assurance, or commissioning of highly-conformal noncoplanar radiotherapy programs.

Radiation Therapy Physics Quality Assurance and Management Practices in Low- and Middle-Income Countries: An Initial Pilot Survey in Six Countries and Validation Through a Site Visit.

Purpose: Our purpose was to assess physics quality assurance (QA) practices in less resourced radiation therapy (RT) centers to improve quality of care. Methods and Materials: A preliminary study was conducted in 2020 of 13 select RT centers in 6 countries, and in 2021, our team conducted onsite visits to all the RT centers in Ghana, one of the countries from the initial survey. The RT centers included 1 private and 2 public institutions (denoted as Public-1 and Public-2). Follow-up surveys were sent to 17 medical physicists from the site visit. Questions centered on the topics of equipment, institutional practice, physics quality assurance, management, and safety practices. Qualitative and descriptive methods were used for data analysis. Questions regarding operational challenges (machine downtime, patient-related issues, power outages, and staffing) were asked on a 5-point Likert scale. Results: The preliminary survey from 2020 had a 92% response rate. One key result showed that for RT centers in lower gross national income per capita countries there was a direct correlation between QA needs and the gross national income per capita of the country. The needs identified included film/array detectors, independent dose calculation software, calibration of ion chambers, diodes, thermoluminiscence diodes (TLDs), phantoms for verification, Treatment Planning System (TPS) test phantoms, imaging test phantoms and film dosimeters, education, and training. For the post survey after the site visit in 2021, we received a 100% response rate. The private and the Public-1 institutions each have computed tomography simulators located in their RT center. The average daily patient external beam workload for each clinic on a linear accelerator was: private = 25, Public-1 = 55, Public-2 = 40. The Co-60 workload was: Public-1 = 45, Public-2 = 25 (there was no Co-60 at the private hospital). Public-1 and -2 lacked the equipment necessary to conform to best practices in Task Group reports (TG) 142 and 198. Public-2 reported significant operational challenges. Notably, Public-1 and -2 have peer review chart rounds, which are attended by clinical oncologists, medical physicists, physicians, and physics trainees. All 17 physicists who responded to the post site visit survey indicated they had a system of documenting, tracking, and trending patient-related safety incidents, but only 1 physicist reported using International Atomic Energy Agency Safety in Radiation Oncology. Conclusions: The preliminary study showed a direct correlation between QA needs and the development index of a country, and the follow-up survey examines operational and physics QA practices in the RT clinics in Ghana, one of the initial countries surveyed. This will form the basis of a planned continent-wide survey in Africa intended to spotlight QA practices in low- and middle-income countries, the challenges faced, and lessons learned to help understand the gaps and needs to support local physics QA and management programs. Audits during the site visit show education and training remain the most important needs in operating successful QA programs.

Preclinical Ultra-High Dose Rate (FLASH) Proton Radiation Therapy System for Small Animal Studies.

Purpose: Animal studies with ultrahigh dose-rate radiation therapy (FLASH, >40 Gy/s) preferentially spare normal tissues without sacrificing antitumor efficacy compared with conventional dose-rate radiation therapy (CONV). At the University of Washington, we developed a cyclotron-generated preclinical scattered proton beam with FLASH dose rates. We present the technical details of our FLASH radiation system and preliminary biologic results from whole pelvis radiation. Methods and Materials: A Scanditronix MC50 compact cyclotron beamline has been modified to produce a 48.7 MeV proton beam at dose rates between 0.1 and 150 Gy/s. The system produces a 6 cm diameter scattered proton beam (flat to ± 3%) at the target location. Female C57BL/6 mice 5 to 6 weeks old were used for all experiments. To study normal tissue effects in the distal colon, mice were irradiated using the entrance region of the proton beam to the whole pelvis, 18.5 Gy at different dose rates: control, CONV (0.6-1 Gy/s) and FLASH (50-80 Gy/s). Survival was monitored daily and EdU (5-ethynyl-2´-deoxyuridine) staining was performed at 24- and 96-hours postradiation. Cleaved caspase-3 staining was performed 24-hours postradiation. To study tumor control, allograft B16F10 tumors were implanted in the right flank and received 18 Gy CONV or FLASH proton radiation. Tumor growth and survival were monitored. Results: After 18.5 Gy whole pelvis radiation, survival was 100% in the control group, 0% in the CONV group, and 44% in the FLASH group (P < .01). EdU staining showed cell proliferation was significantly higher in the FLASH versus CONV group at both 24-hours and 96-hours postradiation in the distal colon, although both radiation groups showed decreased proliferation compared with controls (P < .05). Lower cleaved caspase-3 staining was seen in the FLASH versus conventional group postradiation (P < .05). Comparable flank tumor control was observed in the CONV and FLASH groups. Conclusions: We present our preclinical FLASH proton radiation system and biologic results showing improved survival after whole pelvis radiation, with equivalent tumor control.

A deep-learning-based dose verification tool utilizing fluence maps for a cobalt-60 compensator-based intensity-modulated radiation therapy system.

Background and purpose: A novel cobalt-60 compensator-based intensity-modulated radiation therapy (IMRT) system was developed for a resource-limited environment but lacked an efficient dose verification algorithm. The aim of this study was to develop a deep-learning-based dose verification algorithm for accurate and rapid dose predictions. Materials and methods: A deep-learning network was employed to predict the doses from static fields related to beam commissioning. Inputs were a cube-shaped phantom, a beam binary mask, and an intersecting volume of the phantom and beam binary mask, while output was a 3-dimensional (3D) dose. The same network was extended to predict patient-specific doses for head and neck cancers using two different approaches. A field-based method predicted doses for each field and combined all calculated doses into a plan, while the plan-based method combined all nine fluences into a plan to predict doses. Inputs included patient computed tomography (CT) scans, binary beam masks, and fluence maps truncated to the patient's CT in 3D. Results: For static fields, predictions agreed well with ground truths with average deviations of less than 0.5% for percent depth doses and profiles. Even though the field-based method showed excellent prediction performance for each field, the plan-based method showed better agreement between clinical and predicted dose distributions. The distributed dose deviations for all planned target volumes and organs at risk were within 1.3 Gy. The calculation speed for each case was within two seconds. Conclusions: A deep-learning-based dose verification tool can accurately and rapidly predict doses for a novel cobalt-60 compensator-based IMRT system.

Nearest Neighbor-Based Strategy to Optimize Multi-View Triplet Network for Classification of Small-Sample Medical Imaging Data.

Multi-view classification with limited sample size and data augmentation is a very common machine learning (ML) problem in medicine. With limited data, a triplet network approach for two-stage representation learning has been proposed. However, effective training and verifying the features from the representation network for their suitability in subsequent classifiers are still unsolved problems. Although typical distance-based metrics for the training capture the overall class separability of the features, the performance according to these metrics does not always lead to an optimal classification. Consequently, an exhaustive tuning with all feature-classifier combinations is required to search for the best end result. To overcome this challenge, we developed a novel nearest-neighbor (NN) validation strategy based on the triplet metric. This strategy is supported by a theoretical foundation to provide the best selection of the features with a lower bound of the highest end performance. The proposed strategy is a transparent approach to identify whether to improve the features or the classifier. This avoids the need for repeated tuning. Our evaluations on real-world medical imaging tasks (i.e., radiation therapy delivery error prediction and sarcoma survival prediction) show that our strategy is superior to other common deep representation learning baselines [i.e., autoencoder (AE) and softmax]. The strategy addresses the issue of feature's interpretability which enables more holistic feature creation such that the medical experts can focus on specifying relevant data as opposed to tedious feature engineering.

Commissioning and dosimetric validation of a novel compensator-based Co-60 IMRT system for evaluating suitability to automated treatment planning.

PURPOSE: A novel compensator-based system has been proposed which delivers intensity-modulated radiation therapy (IMRT) with cobalt-60 beams. This could improve access to advanced radiotherapy in low- and middle-income countries. For this system to be clinically viable and to be adapted into the Radiation Planning Assistant (RPA), being developed to offer automated planning services in low- and middle-income countries, it is necessary to commission and validate it in a commercial treatment planning system (TPS). METHODS: The novel treatment device considered here employs a cobalt-60 source and nine compensators. Each compensator is produced by 3-D printing a thin plastic mold which is then filled on-demand within the machine with reusable 2-mm-diameter spherical tungsten balls. This system was commissioned in the Eclipse TPS and validation tests were conducted with Monte Carlo using Geant4 Application for Tomographic Emission for percentage depth dose, in-plane profiles, penumbra, and IMRT dose validation. And the American Association of Physicists in Medicine Task Group 119 benchmarking testing was performed. Additionally, compensator-based cobalt-60 IMRT plans were created for 46 head-and-neck cancer cases and compared to the linac-based volumetric modulated arc therapy (VMAT) plans used clinically, then dosimetric parameters were evaluated. Beam-on time for each field was calculated. In addition, the measurement was also performed in a limited environment and compared with the Monte Carlo simulations. RESULTS: The differences in percent depth doses and in-plane profiles between the Eclipse and Monte Carlo simulations were 0.65% ± 0.41% and 1.02% ± 0.99%, respectively, and the 80%-20% penumbra agreed within 0.46 ± 0.27 mm. For the Task Group 119 validation plans, all treatment planning goals were met and gamma passing rates were >95% (3%/3 mm criteria). In 46 clinical head-and-neck cases, the cobalt-60 compensator-based IMRT plans had planning target volume (PTV) coverages similar to linac-based VMAT plans: all dosimetric values for PTV were within 1.5%. The organs at risk dose parameters were somewhat higher in cobalt-60 compensator-based IMRT plans versus linac-based VMAT plans. The mean dose differences for the spinal cord, brain, and brainstem were 4.43 ± 1.92, 3.39 ± 4.67, and 2.40 ± 3.71 Gy, while those for the rest of the organs were <1 Gy. The average beam-on time per field was 0.42 ± 0.10 min for the 6 MV multi-leaf-collimator plans while those for the cobalt-60 compensator plans were 0.17 ± 0.01 and 0.31 ± 0.01 min at the dose rates of 350 and 175 cGy/min. There was a good agreement between in-plane profiles from measurements and Monte Carlo simulations, which differences are 1.34 ± 1.90% and 0.13 ± 2.16% for two different fields. CONCLUSIONS: A novel compensator-based IMRT system using cobalt-60 beams was commissioned and validated in a commercial TPS. Plan quality with this system was comparable to that of linac-based plans in all test cases with shorter estimated beam-on times. This system enables reliable, high-quality plans with reduced cost and complexity and may have benefits for underserved regions of the world. This system is being integrated into the RPA, a web-based platform for auto-contouring and auto-planning.

Prevention of a wrong-location misadministration through the use of an intradepartmental incident learning system.

PURPOSE: A series of examples are presented in which potential errors in the delivery of radiation therapy were prevented through use of incident learning. These examples underscore the value of reporting near miss incidents. METHODS: Using a departmental incident learning system, eight incidents were noted over a two-year period in which fields were treated "out-of-sequence," that is, fields from a boost phase were treated, while the patient was still in the initial phase of treatment. As a result, an error-prevention policy was instituted in which radiation treatment fields are "hidden" within the oncology information system (OIS) when they are not in current use. In this way, fields are only available to be treated in the intended sequence and, importantly, old fields cannot be activated at the linear accelerator control console. RESULTS: No out-of-sequence treatments have been reported in more than two years since the policy change. Furthermore, at least three near-miss incidents were detected and corrected as a result of the policy change. In the first two, the policy operated as intended to directly prevent an error in field scheduling. In the third near-miss, the policy operated "off target" to prevent a type of error scenario that it was not directly intended to prevent. In this incident, an incorrect digitally reconstructed radiograph (DRR) was scheduled in the OIS for a patient receiving lung cancer treatment. The incorrect DRR had an isocenter which was misplaced by approximately two centimeters. The error was a result of a field from an old plan being scheduled instead of the intended new plan. As a result of the policy described above, the DRR field could not be activated for treatment however and the error was discovered and corrected. Other quality control barriers in place would have been unlikely to have detected this error. CONCLUSIONS: In these examples, a policy was adopted based on incident learning, which prevented several errors, at least one of which was potentially severe. These examples underscore the need for a rigorous, systematic incident learning process within each clinic. The experiences reported in this technical note demonstrate the value of near-miss incident reporting to improve patient safety.

Hazardous Attitudes: Physician Decision Making in Radiation Oncology.

Purpose: The Federal Aviation Administration quantifies hazardous attitudes (HAs) among pilots using a scale. HAs have been linked to aviation risk. We assessed the influence of HAs and other factors in treatment decision making in radiation oncology (RO). Methods and Materials: An anonymous survey was sent to 809 radiation oncologists in US cities housing the top 25 cancer centers. The survey included an HA scale adapted for RO and presented 9 cases assessing risk-tolerant radiation therapy prescribing habits and compliance with the American Society for Radiation Oncology's Choosing Wisely recommendations. Demographic and treatment decision data were dichotomized to identify factors associated with prescribing habits using univariable and multivariable (MVA) logistic regression analyses. Results: A total of 139 responses (17.1%) were received, and 103 were eligible for analysis. Among respondents, 40% were female, ages were evenly distributed, and 83% were in academics. Median scores for all attitudes (macho, anti-authority, worry, resignation, and impulsivity) were below the aviation thresholds for hazard and data from surgical specialties. On MVA, responders >50 years old with >5 years' experience were 4.45 times more likely to recommend risk-tolerant radiation (P = .016). Macho attitude was negatively associated with Choosing Wisely compliant treatments (odds ratio [OR], 0.12; P = .001). Physicians who reported having previously retreated the supraclavicular fossa without complication were more likely to recommend retreatment in medically unfit patients if they felt the complication was avoided owing to careful planning (OR, 5.2; P = .008). Conclusions: To our knowledge, this represents the first study analyzing physician attitudes in RO and their effect on self-reported treatment decisions. This work suggests that attitude may be among the factors that influence risk-tolerant prescribing practices and compliance with Choosing Wisely recommendations.

A radiotherapy technique to limit dose to neural progenitor cell niches without compromising tumor coverage.

Radiation therapy (RT) for brain tumors is associated with neurocognitive toxicity which may be a result of damage to neural progenitor cells (NPCs). We present a novel technique to limit the radiation dose to NPC without compromising tumor coverage. A study was performed in mice to examine the rationale and another was conducted in humans to determine its feasibility. C57BL/6 mice received localized radiation using a dedicated animal irradiation system with on-board CT imaging with either: (1) Radiation which spared NPC containing regions; (2) Radiation which did not spare these niches; or (3) Sham irradiation. Mice were sacrificed 24 h later and the brains were processed for immunohistochemical Ki-67 staining. For the human component of the study, 33 patients with primary brain tumors were evaluated. Two intensity modulated radiotherapy (IMRT) plans were retrospectively compared: a standard clinical plan and a plan which spares NPC regions while maintaining the same dose coverage of the tumor. The change in radiation dose to the contralateral NPC-containing regions was recorded. In the mouse model, non-NPC-sparing radiation treatment resulted in a significant decrease in the number of Ki67(+) cells in dentate gyrus (DG) (P = 0.008) and subventricular zone (SVZ) (P = 0.005) compared to NPC-sparing radiation treatment. In NPC-sparing clinical plans, NPC regions received significantly lower radiation dose with no clinically relevant changes in tumor coverage. This novel radiation technique should significantly reduce radiation doses to NPC containing regions of the brain which may reduce neurocognitive deficits following RT for brain tumors.

Affordable compensator for IMRT delivery designed for low-and-middle income countries, a Monte Carlo study.

We propose a novel cost-effective compensator that can be used to facilitate access to IMRT in low-and-middle income countries. The compensator has the advantages of simplicity, less downtime, increased reliability and less impact on treatment quality from patient motion during treatment. Moreover, the system can be used with either a cobalt-60 unit or linear accelerator. In this Monte Carlo study, the dosimetric properties of the new compensator design have been evaluated. Results were obtained for different field sizes of cobalt teletherapy machine, and the dose was scored at 0.5 cm depth in a water phantom. The effects of compensator thickness, filling material type and shape, and field size were identified. Furthermore, the percentage depth dose and beam profiles for various field sizes and at different depths were obtained. Beam profiles show no significant signature of the beads relative to a solid compensator; in addition, they exhibit a better flatness while preserving symmetry for all field sizes. A reusable bead-based compensator appears to be feasible, and provides dose distribution similar to a solid compensator with low cost and no hazards. Our results avail the ongoing efforts to expand the reach to IMRT in low- and middle-income countries.

A method for optimizing LINAC treatment geometry for volumetric modulated arc therapy of multiple brain metastases.

PURPOSE: Volumetric modulated arc therapy (VMAT) is a rotational delivery technique in which MLC shapes, dose rate, and gantry rotation speed are optimized to produce conformal dose distributions. The aim of this work is to develop a beam projection method for deriving the optimal table and collimator angles for multilesion treatment planning. METHODS: The method consists of four steps. The first step is to define the vector of beam-eye-view (BEV)-Y-axis in the treatment planning CT coordinates. The second step is to project each target onto the BEV-Y-axis vector. In the third step, the best table and collimator angle are found with a brute-force optimization technique that minimizes MLC leaf sharing between lesions. The fourth step is to generate an optimized VMAT plan with appropriate table/collimator angles and evaluate the plan quality. RESULTS: The authors tested the method on three example cases with targets of various locations in the brain and sizes ranging from 1.18 to 17.86 cm(3). Applying the optimized geometric parameter to generate VMAT plan, a reduction of the 12 Gy volume was more than 6.1% for all cases; the plan homogeneity (D2%-D95%) was improved from 5.88 +/- 1.21 to 5.21 +/- 0.93 Gy vs a VMAT plan with the manufacturer recommended table and collimator angles. CONCLUSIONS: The authors conclude that the use of the projection method minimizes the sharing of MLC leaves between lesions and improves the plan quality for multilesion VMAT delivery.

Stereotactic radiosurgery for spine tumors: review of current literature.

Stereotactic radiosurgery (SR) is increasingly utilized for the treatment of intracranial and extracranial pathology. It is considered an important adjuvant to surgery, chemotherapy or fractionated radiotherapy, and the role of SR as a primary treatment modality continues to be explored. Although SR for spinal lesions is in its infancy, there is a growing body of literature supporting its efficacy. The purpose of this review is to summarize the pertinent literature regarding the use of SR for lesions of the spine and spinal cord. Particular emphasis will be placed on large clinical series of both primary and secondary spine tumors.

A Radiation Oncology-Specific Automated Trigger Indicator Tool for High-Risk, Near-Miss Safety Events.

PURPOSE: Error detection in radiation oncology relies heavily on voluntary reporting, and many adverse events and near misses likely go undetected. Trigger tools use existing data in patient charts to identify otherwise-unaccounted-for events and have been successfully employed in other areas of medicine. We developed an automated radiation oncology-specific trigger tool and validated it against near-miss data from a high-volume incident learning system (ILS). METHODS AND MATERIALS: Twenty triggers were derived from an electronic radiation oncology information system. Data from the systems over an approximately 3.5-year period were split randomly into training and test sets. The probability of a high-grade (grade 3-4) near miss for each treatment course in the training set was estimated using a regularized logistic regression model. The predictive model was applied to the test set. Records for 25 flagged treatment courses with an ILS entry were reviewed to explore the association between triggers and near misses, and 25 flagged courses without an ILS entry were reviewed to detect unreported near misses. RESULTS: Of the 3159 treatment courses analyzed, 357 had a grade 3 to 4 ILS entry; 2210 courses composed the training set, and the test set had 949 courses. Areas under the curve on the training and test sets were 0.650 and 0.652, respectively. Of 20 triggers, 9 reached statistical significance on univariate analysis. Fifty percent of the 25 treatment courses in the test set with the highest predicted likelihood of a high-grade near miss with an ILS entry had a direct relationship between the triggers and the near miss. Review of the 25 treatment courses with the highest predicted likelihood of high-grade near miss without an ILS entry found 2 unreported near-miss events. CONCLUSIONS: The radiation oncology-specific automated trigger tool performed modestly and identified additional treatment courses with near-miss events. Radiation oncology trigger tools deserve further exploration.

Low-Dose Image-Guided Pediatric CNS Radiation Therapy: Final Analysis From a Prospective Low-Dose Cone-Beam CT Protocol From a Multinational Pediatrics Consortium.

BACKGROUND: Lower-dose cone-beam computed tomography protocols for image-guided radiotherapy may permit target localization while minimizing radiation exposure. We prospectively evaluated a lower-dose cone-beam protocol for central nervous system image-guided radiotherapy across a multinational pediatrics consortium. METHODS: Seven institutions prospectively employed a lower-dose cone-beam computed tomography central nervous system protocol (weighted average dose 0.7 mGy) for patients ≤21 years. Treatment table shifts between setup with surface lasers versus cone-beam computed tomography were used to approximate setup accuracy, and vector magnitudes for these shifts were calculated. Setup group mean, interpatient, interinstitution, and random error were estimated, and clinical factors were compared by mixed linear modeling. RESULTS: Among 96 patients, with 2179 pretreatment cone-beam computed tomography acquisitions, median age was 9 years (1-20). Setup parameters were 3.13, 3.02, 1.64, and 1.48 mm for vector magnitude group mean, interpatient, interinstitution, and random error, respectively. On multivariable analysis, there were no significant differences in mean vector magnitude by age, gender, performance status, target location, extent of resection, chemotherapy, or steroid or anesthesia use. Providers rated >99% of images as adequate or better for target localization. CONCLUSIONS: A lower-dose cone-beam computed tomography protocol demonstrated table shift vector magnitude that approximate clinical target volume/planning target volume expansions used in central nervous system radiotherapy. There were no significant clinical predictors of setup accuracy identified, supporting use of this lower-dose cone-beam computed tomography protocol across a diverse pediatric population with brain tumors.

Deep learning for patient-specific quality assurance: Identifying errors in radiotherapy delivery by radiomic analysis of gamma images with convolutional neural networks.

PURPOSE: Patient-specific quality assurance (QA) for intensity-modulated radiation therapy (IMRT) is a ubiquitous clinical procedure, but conventional methods have often been criticized as being insensitive to errors or less effective than other common physics checks. Recently, there has been interest in the application of radiomics, quantitative extraction of image features, to radiotherapy QA. In this work, we investigate a deep learning approach to classify the presence or absence of introduced radiotherapy treatment delivery errors from patient-specific QA. METHODS: Planar dose maps from 186 IMRT beams from 23 IMRT plans were evaluated. Each plan was transferred to a cylindrical phantom CT geometry. Three sets of planar doses were exported from each plan corresponding to (a) the error-free case, (b) a random multileaf collimator (MLC) error case, and (c) a systematic MLC error case. Each plan was delivered to the electronic portal imaging device (EPID), and planned and measured doses were used to calculate gamma images in an EPID dosimetry software package (for a total of 558 gamma images). Two radiomic approaches were used. In the first, a convolutional neural network with triplet learning was used to extract image features from the gamma images. In the second, a handcrafted approach using texture features was used. The resulting metrics from both approaches were input into four machine learning classifiers (support vector machines, multilayer perceptrons, decision trees, and k-nearest-neighbors) in order to determine whether images contained the introduced errors. Two experiments were considered: the two-class experiment classified images as error-free or containing any MLC error, and the three-class experiment classified images as error-free, containing a random MLC error, or containing a systematic MLC error. Additionally, threshold-based passing criteria were calculated for comparison. RESULTS: In total, 303 gamma images were used for model training and 255 images were used for model testing. The highest classification accuracy was achieved with the deep learning approach, with a maximum accuracy of 77.3% in the two-class experiment and 64.3% in the three-class experiment. The performance of the handcrafted approach with texture features was lower, with a maximum accuracy of 66.3% in the two-class experiment and 53.7% in the three-class experiment. Variability between the results of the four machine learning classifiers was lower for the deep learning approach vs the texture feature approach. Both radiomic approaches were superior to threshold-based passing criteria. CONCLUSIONS: Deep learning with convolutional neural networks can be used to classify the presence or absence of introduced radiotherapy treatment delivery errors from patient-specific gamma images. The performance of the deep learning network was superior to a handcrafted approach with texture features, and both radiomic approaches were better than threshold-based passing criteria. The results suggest that radiomic QA is a promising direction for clinical radiotherapy.

Characterization of a Bayesian network-based radiotherapy plan verification model.

PURPOSE: The current process for radiotherapy treatment plan quality assurance relies on human inspection of treatment plans, which is time-consuming, error prone and oft reliant on inconsistently applied professional judgments. A previous proof-of-principle paper describes the use of a Bayesian network (BN) to aid in this process. This work studied how such a BN could be expanded and trained to better represent clinical practice. METHODS: We obtained 51 540 unique radiotherapy cases including diagnostic, prescription, plan/beam, and therapy setup factors from a de-identified Elekta oncology information system from the years 2010-2017 from a single institution. Using a knowledge base derived from clinical experience, factors were coordinated into a 29-node, 40-edge BN representing dependencies among the variables. Conditional probabilities were machine learned using expectation maximization module using all data except a subset of 500 patient cases withheld for testing. Different classes of errors that were obtained from incident learning systems were introduced to the testing set of cases which were withheld from the dataset used for building the BN. Different sizes of datasets were used to train the network. In addition, the BN was trained using data from different length epochs as well as different eras. Its performance under these different conditions was evaluated by means of Areas Under the receiver operating characteristic Curve (AUC). RESULTS: Our performance analysis found AUCs of 0.82, 0.85, 0.89, and 0.88 in networks trained with 2-yr, 3-yr 4-yr and 5-yr windows. With a 4-yr sliding window, we found AUC reduction of 3% per year when moving the window back in time in 1-yr steps. Compared to the 4-yr window moved back by 4 yrs (2010-2013 vs 2014-2017), the largest component of overall reduction in AUC over time was from the loss of detection performance in plan/beam error types. CONCLUSIONS: The expanded BN method demonstrates the ability to detect classes of errors commonly encountered in radiotherapy planning. The results suggest that a 4-yr training dataset optimizes the performance of the network in this institutional dataset, and that yearly updates are sufficient to capture the evolution of clinical practice and maintain fidelity.