ARTInp: CBCT-to-CT Image Inpainting and Image Translation in Radiotherapy

TL;DR

The paper tackles the challenge of CBCT limitations in Adaptive Radiation Therapy, especially for whole-body treatments like Total Marrow and Lymph Node Irradiation, by integrating CBCT gap inpainting with CBCT-to-CT translation to produce synthetic CTs (sCTs) for treatment validation. It introduces ARTInp, a dual-GAN framework comprising a completion network for CBCT gap filling and a translation network to generate sCT from inpainted CBCT, trained and evaluated on the SynthRad 2023 brain CBCT/CT dataset with artificial gaps. Quantitatively, ARTInp achieves MAE$\%$ below $2.5\%$, PSNR around $27$ dB, and SSIM around $0.77$–$0.79$, indicating feasible improvement of CBCT-based workflows though with residual artifacts and SSIM limitations. The work provides a reproducible prototype and discusses future directions including joint training, 3D or diffusion-based approaches, and clinical evaluation in settings such as TMLI to fully assess impact on ART accuracy and safety.

Abstract

A key step in Adaptive Radiation Therapy (ART) workflows is the evaluation of the patient's anatomy at treatment time to ensure the accuracy of the delivery. To this end, Cone Beam Computerized Tomography (CBCT) is widely used being cost-effective and easy to integrate into the treatment process. Nonetheless, CBCT images have lower resolution and more artifacts than CT scans, making them less reliable for precise treatment validation. Moreover, in complex treatments such as Total Marrow and Lymph Node Irradiation (TMLI), where full-body visualization of the patient is critical for accurate dose delivery, the CBCT images are often discontinuous, leaving gaps that could contain relevant anatomical information. To address these limitations, we propose ARTInp (Adaptive Radiation Therapy Inpainting), a novel deep-learning framework combining image inpainting and CBCT-to-CT translation. ARTInp employs a dual-network approach: a completion network that fills anatomical gaps in CBCT volumes and a custom Generative Adversarial Network (GAN) to generate high-quality synthetic CT (sCT) images. We trained ARTInp on a dataset of paired CBCT and CT images from the SynthRad 2023 challenge, and the performance achieved on a test set of 18 patients demonstrates its potential for enhancing CBCT-based workflows in radiotherapy.

Figures (6)

• Figure 1: An example of the usage of CBCT before a TMLI session: coronal and sagittal views of several series of CBCT scans combined together to cover the whole body of the patient.
• Figure 2: The overview of the ARTInp framework that includes a completion network and a translation network.
• Figure 3: The architecture of the completion network.
• Figure 4: The architecture of the translation network that uses the axial CBCT slices in the 16-bit change and outputs sCT slices.
• Figure 5: Examples of the sCT generation.