Quality assurance for online adaptive radiotherapy: a secondary dose verification model with geometry-encoded U-Net

Abstract

In online adaptive radiotherapy (ART), quick computation-based secondary dose verification is crucial for ensuring the quality of ART plans while the patient is positioned on the treatment couch. However, traditional dose verification algorithms are generally time-consuming, reducing the efficiency of ART workflow. This study aims to develop an ultra-fast deep-learning (DL) based secondary dose verification algorithm to accurately estimate dose distributions using computed tomography (CT) and fluence maps (FMs). We integrated FMs into the CT image domain by explicitly resolving the geometry of treatment delivery. For each gantry angle, an FM was constructed based on the optimized multi-leaf collimator apertures and corresponding monitoring units. To effectively encode treatment beam configuration, the constructed FMs were back-projected to $30$ cm away from the isocenter with respect to the exact geometry of the treatment machines. Then, a 3D U-Net was utilized to take the integrated CT and FM volume as input to estimate dose. Training and validation were performed on $381$ prostate cancer cases, with an additional $40$ testing cases for independent evaluation of model performance. The proposed model can estimate dose in ∼ $15$ ms for each patient. The average γ passing rate ( $3\%/2\text{mm}$ , $10\%$ threshold) for the estimated dose was 99.9% ± 0.15% on testing patients. The mean dose differences for the planning target volume and organs at risk were $0.07\%\pm 0.34\%$ and $0.48\%\pm 0.72\%$ , respectively. We have developed a geometry-resolved DL framework for accurate dose estimation and demonstrated its potential in real-time online ART doses verification.

Keywords: artificial intelligence; deep learning; prostate cancer; real-time calculation; secondary dose verification.

Figure 1.

Clinical secondary dose verification workflow according to TG-219 guidelines [8] (upper route) vs proposed DL-based secondary dose verification workflow by using an integrated volume encoding FMs into CT domain as input for 3D U-Net (lower route).

Figure 2.

(a) Illustration of encoding FMs to the patient CT image domain. (b) Schematic diagram of proposed 3D U-Net module.

Figure 3.

Training and validation loss of the proposed model along training epochs.

Figure 4.

Model performance on a representative testing case. (a)–(c) Ground truth dose distribution, estimated dose distribution, and their difference map (display window $\left[-3\mathrm{\%},3\mathrm{\%}\right]$ of Rx. i.e. 45 Gy) with PTV (green), bladder (red), rectum (blue), and femoral heads (yellow) contours; (d) DVH plots of ground truth (solid) and predicted dose (dashed).

Figure 5.

Boxplot comparisons of dose differences for and in PTV, bladder, rectum, and femoral heads.