Prospective risk analysis of the online-adaptive artificial intelligence-driven workflow using the Ethos treatment system

Multi-institutional questionnaire-based survey on online adaptive radiotherapy performed using commercial systems in Japan in 2023

In this study, we aimed to conduct a survey on the current clinical practice of, staffing for, commissioning of, and staff training for online adaptive radiotherapy (oART) in the institutions that installed commercial oART systems in Japan, and to share the information with institutions that will implement oART systems in future. A web-based questionnaire, containing 107 questions, was distributed to nine institutions in Japan. Data were collected from November to December 2023. Three institutions each with the MRIdian (ViewRay, Oakwood Village, OH, USA), Unity (Elekta AB, Stockholm, Sweden), and Ethos (Varian Medical Systems, Palo Alto, CA, USA) systems completed the questionnaire. One institution (MRIdian) had not performed oART by the response deadline. Each institution had installed only one oART system. Hypofractionation, and moderate hypofractionation or conventional fractionation were employed in the MRIdian/Unity and Ethos systems, respectively. The elapsed time for the oART process was faster with the Ethos than with the other systems. All institutions added additional staff for oART. Commissioning periods differed among the oART systems owing to provision of beam data from the vendors. Chambers used during commissioning measurements differed among the institutions. Institutional training was provided by all nine institutions. To the best of our knowledge, this was the first survey about oART performed using commercial systems in Japan. We believe that this study will provide useful information to institutions that installed, are installing, or are planning to install oART systems.

Safety and Efficiency Analysis of Operational Decision-Making During Cone Beam Computed Tomography-Based Online Adaptive Radiation Therapy

PURPOSE

Cone beam computed tomography (CBCT)-based online adaptive radiation therapy (ART) is especially beneficial for patients with large interfractional anatomic changes. However, treatment planning and review decisions need to be made at the treatment console in real-time and may be delegated to clinical staff whose conventional scope of practice does not include making such decisions. Therefore, implementation can create new safety risks and inefficiencies. The objective of this work is to systematically analyze the safety and efficiency implications of human decision-making during the treatment session for CBCT-based online ART.

METHODS AND MATERIALS

The analysis was performed by applying the Systems-Theoretical Process Analysis technique and its extension for human decision-making. Four centers of different CBCT-based online ART practice models comprised the analysis team.

RESULTS

The general radiation therapy control structure was refined to model the interactions between routine treatment delivery staff and in-person or remote support staff. The treatment delivery staff perform 6 key control actions. Eighteen undesirable states of those control actions were identified as affecting safety and/or efficiency. In turn, 97 hazardous clinical scenarios were identified, with the control action "prepare and position patient" having the least number of scenarios and "delineate/edit influencer and target structures" having the most. Five of these are specific to either in-person or remote support during the treatment session, and 12 arise from staff support in general.

CONCLUSIONS

An optimally safe and efficient online ART program should require little to no support staff at the treatment console to reduce staff coordination. Uptraining of the staff already at the treatment console is needed to achieve this goal. Beyond the essential knowledge and skills such as contour editing and the selection of an optimal plan, uptraining should also target the specific cognitive biases identified in this work and the cognitive strategies to overcome these biases. Additionally, technological and organizational changes are necessary.

Enhancing Safety in AI-Driven Cone Beam CT-based Online Adaptive Radiation Therapy: Development and Implementation of an Interdisciplinary Workflow

Purpose

The emerging online adaptive radiation therapy (OART) treatment strategy based on cone beam computed tomography allows for real-time replanning according to a patient's current anatomy. However, implementing this procedure requires a new approach across the patient's care path and monitoring of the "black box" adaptation process. This study identifies high-risk failure modes (FMs) associated with AI-driven OART and proposes an interdisciplinary workflow to mitigate potential medical errors from highly automated processes, enhance treatment efficiency, and reduce the burden on clinicians.

Methods and Materials

An interdisciplinary working group was formed to identify safety concerns in each process step using failure mode and effects analysis (FMEA). Based on the FMEA results, the team designed standardized procedures and safety checklists to prevent errors and ensure successful task completion. The Risk Priority Numbers (RPNs) for the top twenty FMs were calculated before and after implementing the proposed workflow to evaluate its effectiveness. Three hundred seventy-four adaptive sessions across 5 treatment sites were performed, and each session was evaluated for treatment safety and FMEA assessment.

Results

The OART workflow has 4 components, each with 4, 8, 13, and 4 sequentially executed tasks and safety checklists. Site-specific template preparation, which includes disease-specific physician directives and Intelligent Optimization Engine template testing, is one of the new procedures introduced. The interdisciplinary workflow significantly reduced the RPNs of the high-risk FMs, with an average decrease of 110 (maximum reduction of 305.5 and minimum reduction of 27.4).

Conclusions

This study underscores the importance of addressing high-risk FMs associated with AI-driven OART and emphasizes the significance of safety measures in its implementation. By proposing a structured interdisciplinary workflow and integrated checklists, the study provides valuable insights into ensuring the safe and efficient delivery of OART while facilitating its effective integration into clinical practice.

Evaluation of the Ethos synthetic computed tomography for bolus-covered surfaces

PURPOSE

Ethos allows online adaption of radiotherapy treatment plans. Dose is calculated on synthetic computed tomographies (sCT), CT-like images generated by deforming planning CTs (pCT) onto daily cone beam CTs (CBCT) acquired during treatment sessions. Errors in sCT density distribution may lead to dose calculation errors. sCT correctness was investigated for bolus-covered surfaces.

METHODS

pCTs were recorded of a slab phantom covered with bolus of different thicknesses and with air gaps introduced by spacer rings of variable diameters and heights. Treatment plans were irradiated following the adaptive workflow with different bolus configurations present in the pCT and CBCT. sCT densities were compared to those of the pCT for the same air gap size. Additionally, the neck region of an anthropomorphic phantom was imaged using a plane standard bolus versus an individual bolus adapted to the phantom's outer contour.

RESULTS

Varying bolus thickness by 5 mm between pCT and CBCT was reproduced in the sCT within 2 mm accuracy. Different air gaps in pCT and CBCT resulted in highly variable bolus thickness in the sCT with a typical error of 5 mm or more. In extreme cases, air gaps were filled with bolus material density in the sCT or the phantom was unrealistically deformed near changed bolus geometries. Changes in bolus thickness and deformation also occurred in the anthropomorphic phantom.

CONCLUSION

sCTs must be critically examined and included in plan-specific quality assurance. The use of tight-fitting air gap-free bolus should be preferred to increase the similarity between sCT and CBCT.