Domain Adaptation of Automated Treatment Planning from Computed Tomography to Magnetic Resonance

TL;DR

This study demonstrates that machine learning models trained on CT images can be adapted to generate acceptable treatment plans on MR images, with some dose differences influenced by imaging and anatomical variations.

Contribution

The paper shows the feasibility of applying CT-trained ML treatment planning models to MR images through domain adaptation, addressing a key challenge in MR-only workflows.

Findings

MR plans met 93.1% of evaluation criteria, close to CT plans at 96.3%.

About half of the dosimetric differences are due to imaging modality changes.

Anatomical differences significantly impact dose deviations.

Abstract

Objective: Machine learning (ML) based radiation treatment (RT) planning addresses the iterative and time-consuming nature of conventional inverse planning. Given the rising importance of Magnetic resonance (MR) only treatment planning workflows, we sought to determine if an ML based treatment planning model, trained on computed tomography (CT) imaging, could be applied to MR through domain adaptation. Methods: In this study, MR and CT imaging was collected from 55 prostate cancer patients treated on an MR linear accelerator. ML based plans were generated for each patient on both CT and MR imaging using a commercially available model in RayStation 8B. The dose distributions and acceptance rates of MR and CT based plans were compared using institutional dose-volume evaluation criteria. The dosimetric differences between MR and CT plans were further decomposed into setup, cohort, and imaging domain components. Results: MR plans were highly acceptable, meeting 93.1% of all evaluation criteria compared to 96.3% of CT plans, with dose equivalence for all evaluation criteria except for the bladder wall, penile bulb, small and large bowel, and one rectum wall criteria (p<0.05). Changing the input imaging modality (domain component) only accounted for about half of the dosimetric differences observed between MR and CT plans. Anatomical differences between the ML training set and the MR linac cohort (cohort component) were also a significant contributor. Significance: We were able to create highly acceptable MR based treatment plans using a CT-trained ML model for treatment planning, although clinically significant dose deviations from the CT based plans were observed.