Medical physics practice guideline 4.b: 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 8000 members and is the principal organization of medical physicists in the US. 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 US. 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. While must is the term to be used in the guidelines, if an entity that adopts the guideline has shall as the preferred term, the AAPM considers that must and shall have the same meaning. Should and should not: Used to indicate a prudent practice to which exceptions may occasionally be made in appropriate circumstances.

Practical Implementation of Emergent After-Hours Radiation Treatment Process Using Remote Treatment Planning on Optimized Diagnostic CT Scans.

The purpose of this report is to present the implementation of a process for after-hours radiation treatment (RT) utilizing remote treatment planning based on optimized diagnostic computed tomography (CT) scans for the urgent palliative treatment of inpatients. A standardized operating procedure was developed by an interprofessional panel to improve the quality of after-hours RT and minimize the risk of treatment errors. A new diagnostic CT protocol was created that could be performed after-hours on hospital scanners and would ensure a reproducible patient position and adequate field of view. An on-call structure for dosimetry staff was created utilizing remote treatment planning. The optimized CT protocol was developed in collaboration with the radiology department, and a novel order set was created in the electronic health system. The clinical workflow begins with the radiation oncologist notifying the on-call team (therapist, dosimetrist, and physicist) and obtaining an optimized diagnostic CT scan on a hospital-based scanner. The dosimetrist remotely creates a plan; the physicist checks the plan; and the patient is treated. Plans are intentionally simple (parallel opposed fields, symmetric jaws) to expedite care and reduce the risk of error. Education on the new process was provided for all relevant staff. Our process was successfully implemented with the use of an optimized CT protocol and remote treatment planning. This approach has the potential to improve the quality and safety of emergent after-hours RT by better approximating the normal process of care.

Technical Note: Deep Learning approach for automatic detection and identification of patient positioning devices for radiation therapy.

PURPOSE: Automatic detection and identification of setup devices, using a deep convolutional neural network (CNN) for real-time multiclass object detection, has the potential to reduce errors in the treatment delivery process by avoiding documentation errors. METHODS: A database of the setup device photos from the most recent 1200 patients treated at our institution was downloaded from the record and verify (R&V) system along with the corresponding setup notes. Images were manually labeled with bounding boxes of each device. A real-time object detection CNN using the "you only look once" (YOLOv2) architecture was trained using transfer learning of a pretrained CNN (ResNet50). The CNN was trained to detect and identify 11 of the most common treatment accessories used at our institution. RESULTS: Using transfer learning of a CNN for multiclass object detection, we are able to automatically detect and identify setup devices in photographs with an accuracy of 96%. CONCLUSIONS: Automation in radiation oncology has the potential to reduce risk. Automatic detection of setup devices is possible using a CNN and transfer learning. This work shows both the value of incident learning systems (ILS) in practice knowledge dissemination, and shows how automation of clinical processes and less reliance on manual documentation has the potential for risk reduction in radiation oncology treatments.

Integration of automation into an existing clinical workflow to improve efficiency and reduce errors in the manual treatment planning process for total body irradiation (TBI).

PURPOSE: To identify causes of error, and present the concept of an automated technique that improves efficiency and helps to reduce transcription and manual data entry errors in the treatment planning of total body irradiation (TBI). METHODS: Analysis of incidents submitted to incident learning system (ILS) was performed to identify potential avenues for improvement by implementation of automation of the manual treatment planning process for total body irradiation (TBI). Following this analysis, it became obvious that while the individual components of the TBI treatment planning process were well implemented, the manual 'bridging' of the components (transcribing data, manual data entry etc.) were leading to high potential for error. A C#-based plug-in treatment planning script was developed to remove the manual parts of the treatment planning workflow that were contributing to increased risk. RESULTS: Here we present an example of the implementation of "Glue" programming, combining treatment planning C# scripts with existing spreadsheet calculation worksheets. Prior to the implementation of automation, 35 incident reports related to the TBI treatment process were submitted to the ILS over a 6-year period, with an average of 1.4 ± 1.7 reports submitted per quarter. While no incidents reached patients, reports ranged from minor documentation issues to potential for mistreatment if not caught before delivery. Since the implementation of automated treatment planning and documentation, treatment planning time per patient, including documentation, has been reduced; from an average of 45 min pre-automation to <20 min post-automation. CONCLUSIONS: Manual treatment planning techniques may be well validated, but they are time-intensive and have potential for error. Often the barrier to automating these techniques becomes the time required to "re-code" existing solutions in unfamiliar computer languages. We present the workflow here as a proof of concept that automation may help to improve clinical efficiency and safety for special procedures.

The Current State of Physics Plan Review Training in Medical Physics Residency Programs in North America.

PURPOSE: This study aimed to identify the current state of residency training in physics plan reviews. METHODS AND MATERIALS: A voluntary, anonymous survey was sent to all program directors of accredited therapeutic medical physics residency programs in North America. Survey questions were developed to determine whether and how residents are trained in physics plan reviews. Survey questions were developed using expert validation and cognitive pretesting. RESULTS: Using a prospectively approved study (COMIRB 18-1073), responses were collected from 70 program directors, representing a 70% response rate. All respondents (100%) designated patient safety to be the purpose of physics plan reviews. Of the respondents, 94% indicated that physicists should first receive training in physics plan reviews while in a residency program. The vast majority of respondents (99%) provide training to residents in physics plan reviews. Although 57 programs (81% of respondents) have residents perform physics plan reviews as part of clinical practice (with varying levels of independence), 13 programs (19% of respondents) do not. The majority of respondents use the following training methods: observe staff physicists (96%), perform supervised reviews on actual patients for training or clinical practice (93%), use a checklist (80%), and read reference materials (62%). Although simulation plans with embedded errors would be implemented by 71% of respondents, they are currently used in only 19% of programs. CONCLUSIONS: The present study is the first to characterize chart-check teaching practices in medical physics residency programs. The vast majority of programs currently train residents in physics plan reviews. The most common teaching methods are observing and performing physics plan reviews, but there is variability in the level of resident involvement in clinical practice for physics plan reviews. There is room for the field to consider advancing current training methods, which is especially important given the critical roles that physics plan reviews have with regard to patient safety.

Towards a paperless medical physics residency management system.

Documentation is a required component of a residency program, but can be difficult to collect and disseminate, resulting in minimal utilization by residents and faculty. The purpose of this work is to adapt a commercially-available Web-based medical residency management system to improve the learning experience by efficiently distributing program information, documenting resident activities, and providing frequent monitoring and timely feedback of resident progress. To distribute pro- gram information, program requirements and rotation readings were uploaded. An educational conference calendar was created with associated files and attendance records added. To document resident progress, requirements for over 37 different clinical procedures were added, for which the resident logged the total number of procedures performed. Progress reports were created and automatically distributed. To provide feedback to the resident, an extensive electronic evaluation system was created. Results are shown for the initial 21 months of program existence, consisting of a single resident for the first 12 months and two residents for the subsequent 9 months. The system recorded that 130 documents were uploaded and 100% of required documents were downloaded by the resident. In total, 385 educational conferences and meetings were offered, of which the residents attended 95%. The second-year and first-year residents logged 1030 and 522 clinical proce- dures, respectively. The residents submitted a total of 116 status reports detailing weekly activities, 100% of which were reviewed by faculty within an average of 11.3 days. A total of 65 evaluations of the residents were submitted. The residents reviewed 100% of respective evaluations within an average of 1.5 days. We have successfully incorporated a paperless, Web-based management system in a medical physics residency program. A robust electronic documentation system has been implemented, which has played a central role in enhancing the training experience. 

Dosimetric comparison of left-sided whole breast irradiation with 3DCRT, forward-planned IMRT, inverse-planned IMRT, helical tomotherapy, and topotherapy.

BACKGROUND AND PURPOSE: To compare left-sided whole breast conventional and intensity-modulated radiotherapy (IMRT) treatment planning techniques. MATERIALS AND METHODS: Treatment plans were created for 10 consecutive patients. Three-dimensional conformal radiotherapy (3DCRT), forward-planned IMRT (for-IMRT), and inverse-planned IMRT (inv-IMRT) used two tangent beams. For-IMRT utilized up to four segments per beam. For helical tomotherapy (HT) plans, beamlet entrance and/or exit to critical structures was blocked. Topotherapy plans, which used static gantry angles with simultaneous couch translation and inverse-planned intensity modulation, used two tangent beams. Plans were normalized to 50Gy to 95% of the retracted PTV. RESULTS: Target max doses were reduced with for-IMRT compared to 3DCRT, which were further reduced with HT, topotherapy, and inv-IMRT. HT resulted in lowest heart and ipsilateral lung max doses, but had higher mean doses. Inv-IMRT and topotherapy reduced ipsilateral lung mean and max doses compared to 3DCRT and for-IMRT. CONCLUSIONS: All modalities evaluated provide adequate coverage of the intact breast. HT, topotherapy, and inv-IMRT can reduce high doses to the target and normal tissues, although HT results in increased low doses to large volume of normal tissue. For-IMRT improves target homogeneity compared with 3DCRT, but to a lesser degree than the inverse-planned modalities.

A phantom model demonstration of tomotherapy dose painting delivery, including managed respiratory motion without motion management.

The aim of the study was to demonstrate a potential alternative scenario for accurate dose-painting (non-homogeneous planned dose) delivery at 1 cm beam width with helical tomotherapy (HT) in the presence of 1 cm, three-dimensional, intra-fraction respiratory motion, but without any active motion management. A model dose-painting experiment was planned and delivered to the average position (proper phase of a 4DCT scan) with three spherical PTV levels to approximate dose painting to compensate for hypothetical hypoxia in a model lung tumor. Realistic but regular motion was produced with the Washington University 4D Motion Phantom. A small spherical Virtual Water phantom was used to simulate a moving lung tumor inside of the LUNGMAN anthropomorphic chest phantom to simulate realistic heterogeneity uncertainties. A piece of 4 cm Gafchromic EBT film was inserted into the 6 cm diameter sphere. TomoTherapy, Inc., DQA software was used to verify the delivery performed on a TomoTherapy Hi-Art II device. The dose uncertainty in the purposeful absence of motion management and in the absence of large, low frequency drifts (periods greater than the beam width divided by the couch velocity) or randomness in the breathing displacement yields very favorable results. Instead of interference effects, only small blurring is observed because of the averaging of many breathing cycles and beamlets and the avoidance of interference. Dose painting during respiration with helical tomotherapy is feasible in certain situations without motion management. A simple recommendation is to make respiration as regular as possible without low frequency drifting. The blurring is just small enough to suggest that it may be acceptable to deliver without motion management if the motion is equal to the beam width or smaller (at respiration frequencies) when registered to the average position.

A comprehensive assessment by tumor site of patient setup using daily MVCT imaging from more than 3,800 helical tomotherapy treatments.

PURPOSE: To assess patient setup corrections based on daily megavoltage CT (MVCT) imaging for four anatomic treatment sites treated on tomotherapy. METHOD AND MATERIALS: Translational and rotational setup corrections, based on registration of daily MVCT to planning CT images, were analyzed for 1,179 brain and head and neck (H&N), 1,414 lung, and 1,274 prostate treatment fractions. Frequencies of three-dimensional vector lengths, overall distributions of setup corrections, and patient-specific distributions of random and systematic setup errors were analyzed. RESULTS: Brain and H&N had lower magnitude positioning corrections and smaller variations in translational setup errors but were comparable in roll rotations. Three-dimensional vector translational shifts of larger magnitudes occurred more frequently for lung and prostate than for brain and H&N treatments, yet this was not observed for roll rotations. The global systematic error for prostate was 4.7 mm in the vertical direction, most likely due to couch sag caused by large couch extension distances. Variations in systematic errors and magnitudes of random translational errors ranged from 1.6 to 2.6 mm for brain and H&N and 3.2 to 7.2 mm for lung and prostate, whereas roll rotational errors ranged from 0.8 degrees to 1.2 degrees for brain and H&N and 0.5 degrees to 1.0 degrees for lung and prostate. CONCLUSIONS: Differences in setup were observed between brain, H&N, lung, and prostate treatments. Patient setup can be improved if daily imaging is performed. This analysis can assess the utilization of daily image guidance and allows for further investigation into improved anatomic site-specific and patient-specific treatments.