Abstract
Objective. We propose a semiautomatic pipeline for radiation therapy treatment planning, combining ideas from machine learning–automated planning and multicriteria optimization (MCO). Approach. Using knowledge extracted from historically delivered plans, prediction models for spatial dose and dose statistics are trained and furthermore systematically modified to simulate changes in tradeoff priorities, creating a set of differently biased predictions. Based on the predictions, an MCO problem is subsequently constructed using previously developed dose mimicking functions, designed in such a way that its Pareto surface spans the range of clinically acceptable yet realistically achievable plans as exactly as possible. The result is an algorithm outputting a set of Pareto optimal plans, either fluence-based or machine parameter–based, which the user can navigate between in real time to make adjustments before a final deliverable plan is created. Main results. Numerical experiments performed on a dataset of prostate cancer patients show that one may often navigate to a better plan than one produced by a single-plan-output algorithm. Significance. We demonstrate the potential of merging MCO and a data-driven workflow to automate labor-intensive parts of the treatment planning process while maintaining a certain extent of manual control for the user.
Highlights
Having achieved a broad range of promising results in recent years, the application of machine learning methods to biomedical engineering is today established as a prosperous subject of research (Park et al 2018, Siddique and Chow 2020)
Instead of continuing on the direction of developing a fully automated method, in this paper, we propose a new semiautomatic treatment planning workflow in which the a treatment planner or clinician may optionally articulate their own preferences by navigating in real time between Pareto optimal plans, combining ideas from machine learning and multicriteria optimization (MCO)
While some work has been focused on predicting weights in a weighted-sum objective function (Boutilier et al 2015), the majority of recent work has been aimed at predicting achievable dose-related quantities— for example, spatial dose distributions and dose–volume histograms (DVHs)—and setting up a dose mimicking optimization problem to minimize the deviation from the values evaluated on the actual dose to those predicted
Summary
Having achieved a broad range of promising results in recent years, the application of machine learning methods to biomedical engineering is today established as a prosperous subject of research (Park et al 2018, Siddique and Chow 2020). Within automated treatment planning for radiation therapy, while the direct prediction of machine parameters of an optimal or desired plan remains an intractably high-dimensional and nonlinear problem, data-driven methods based on a prediction–mimicking pipeline has helped in homogenizing the labor-intensive process of creating clinically satisfactory plans (Berry et al 2016). Such a methodology has several fundamental drawbacks—the produced deliverable plan is, for example, highly dependent on the quality of the prediction, and it is in practice often close to, yet not sufficiently in line with, the clinician’s preferences, entailing the need for further post-processing using manual tools (Cagni et al 2018). While some work has been focused on predicting weights in a weighted-sum objective function (Boutilier et al 2015), the majority of recent work has been aimed at predicting achievable dose-related quantities— for example, spatial dose distributions and dose–volume histograms (DVHs)—and setting up a dose mimicking optimization problem to minimize the deviation from the values evaluated on the actual dose to those predicted
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
