Cone beam CT based validation of neural network generated synthetic CTs for radiotherapy in the head region

Abstract

Purpose: In the past years, many different neural network-based conversion techniques for synthesising CTs (sCTs) from MR images have been published. While the model's performance can be checked during the training against the test set, test datasets can never represent the whole population. Conversion errors can still occur for special cases, e.g. for unusual anatomical situations. Therefore, the performance of sCT conversion needs to be verified on a patient specific level, especially in the absence of a planning CT (pCT). In this study, the capability of cone-beam CTs (CBCTs) for the validation of sCTs generated by a neural network was investigated.Methods: 41 patients with tumours in the head region were selected. 20 of them were used for model training and 10 for validation. Different implementations of CycleGAN (with/without identity and feature loss) were used to generate sCTs. The pixel (MAE, RMSE, PSNR) and geometric error (DICE, Sensitivity, Specificity) values were reported to identify the best model. VMAT plans were created for the remaining 11 patients on the pCTs. These plans were re-calculated on sCTs and CBCTs. An automatic density overriding method (CBCT RS ) and a population-based dose calculation method (CBCT P op ) were employed for CBCT based dose calculation. The dose distributions were analysed using 3D global gamma analysis applying a threshold of 10% with respect to the prescribed dose. Differences in DVH metrics for the PTV and the organs at risk were compared among the dose distributions based on pCTs, sCTs, and CBCTs.Results: The best model was the CycleGAN without identity and feature matching loss. Including the identity loss led to a metric decrease of 10% for DICE and a metric increase of 20-60 HU for MAE. Using the 2%/2mm gamma criterion and pCT as reference, the mean gamma pass rates were 99.0±0.4% for sCTs. Mean gamma pass rate values comparing pCT and CBCT were 99.0±0.8% and 99.1±0.8% for the CBCT RS and CBCT P op , respectively. The mean gamma pass rates comparing sCT and CBCT resulted in 98.4±1.6% and 99.2±0.6% for CBCT RS and CBCT P op , respectively. The differences between the gamma-pass-rates of the sCT and two CBCT based methods were not significant. The majority of deviations of the investigated DVH metrices between sCTs and CBCTs were within 2%. Conclusion:The dosimetric results demonstrate good agreement between sCT, CBCT, and pCT based calculations. A properly applied CBCT conversion method can serve as a tool for quality assurance procedures in an MR only radiotherapy workflow for head patients. Dosimetric deviations of DVH metrics between sCT and CBCTs of larger than 2% should be followed-up.

Technical note: First implementation of a one‐stop solution of radiotherapy with full‐workflow automation based on CT‐linac combination

“…However, interest in workflow integration and automation of radiotherapy is growing within the community. Studies on CBCT‐only and MRI‐only planning are underway on large populations and more anatomical sites 29–32 . As a proof‐of‐concept study of one‐stop radiotherapy, the proposed workflow shows a promising prospect with full end‐to‐end automation, and it is expected to migrate and validate it on other mainstream platforms within clinical applicability.…”

Prospective Clinical Feasibility Study for MRI-Only Brain Radiotherapy

“…There is however still a need for case specific QA. One suggestion could be to use CBCT for dose calculation as an independent evaluation of the sCT, which would likely find most clinically relevant deviations (23)(24)(25). Regular sCT generation model QA is another important aspect of implementing a deep learningbased software.…”

Evaluation of a novel CBCT conversion method implemented in a treatment planning system

“…The image quality of CBCT is often lower compared to fan-beam CT and the Houns eld Units (HU) to mass-or electron density conversion cannot be directly converted between these two modalities due to the different amount of scattered radiation and the projection geometry [11,12]. Several groups have investigated different strategies to overcome this limitation of the CBCT images and showed that accurate dose calculation is feasible [13][14][15][16][17][18][19][20][21][22][23].…”