CBCT‐Based synthetic CT image generation using conditional denoising diffusion probabilistic model

Abstract

BackgroundDaily or weekly cone‐beam computed tomography (CBCT) scans are commonly used for accurate patient positioning during the image‐guided radiotherapy (IGRT) process, making it an ideal option for adaptive radiotherapy (ART) replanning. However, the presence of severe artifacts and inaccurate Hounsfield unit (HU) values prevent its use for quantitative applications such as organ segmentation and dose calculation. To enable the clinical practice of online ART, it is crucial to obtain CBCT scans with a quality comparable to that of a CT scan.PurposeThis work aims to develop a conditional diffusion model to perform image translation from the CBCT to the CT distribution for the image quality improvement of CBCT.MethodsThe proposed method is a conditional denoising diffusion probabilistic model (DDPM) that utilizes a time‐embedded U‐net architecture with residual and attention blocks to gradually transform the white Gaussian noise sample to the target CT distribution conditioned on the CBCT. The model was trained on deformed planning CT (dpCT) and CBCT image pairs, and its feasibility was verified in brain patient study and head‐and‐neck (H&N) patient study. The performance of the proposed algorithm was evaluated using mean absolute error (MAE), peak signal‐to‐noise ratio (PSNR) and normalized cross‐correlation (NCC) metrics on generated synthetic CT (sCT) samples. The proposed method was also compared to four other diffusion model‐based sCT generation methods.ResultsIn the brain patient study, the MAE, PSNR, and NCC of the generated sCT were 25.99 HU, 30.49 dB, and 0.99, respectively, compared to 40.63 HU, 27.87 dB, and 0.98 of the CBCT images. In the H&N patient study, the metrics were 32.56 HU, 27.65 dB, 0.98 and 38.99 HU, 27.00, 0.98 for sCT and CBCT, respectively. Compared to the other four diffusion models and one Cycle generative adversarial network (Cycle GAN), the proposed method showed superior results in both visual quality and quantitative analysis.ConclusionsThe proposed conditional DDPM method can generate sCT from CBCT with accurate HU numbers and reduced artifacts, enabling accurate CBCT‐based organ segmentation and dose calculation for online ART.

A retrospective study on the investigation of potential dosimetric benefits of online adaptive proton therapy for head and neck cancer

“…Meanwhile, many groups have been devoting efforts to enhancing the image quality and HU accuracy. 32 Advanced deep learning technologies have been employed to generate synthetic CT images from CBCT images, 33 , 34 presenting an alternative solution for CBCT‐guided online APT. Furthermore, a new on‐board CBCT imaging system named HyperSight (Varian Co., Palo Alto, CA, USA) has been developed for the newer model of the CBCT‐based online adaptive photon therapy system Ethos very recently, which offers CBCT images of significantly improved quality.…”

Diffusion model-based single-scan dual-energy cone-beam CT using primary beam splitting

“…DDPM is one parameterization of diffusion models, which is a category of latent variable models using a Markov chain to convert a white Gaussian distribution to the target data distribution [9]. Diffusion models have demonstrated superior performance in medical imaging tasks, such as medical image reconstruction and medical image synthesis [10][11][12][13][14]. During the forward diffusion process, a sequence of gradually contaminated images 𝑥 % , 𝑥 & , ⋯ , 𝑥 ' are constructed by the distribution 𝑞(𝑥 ( |𝑥 ()% ), which is a normal distribution, as shown in Figure 3 (a).…”

Image-domain material decomposition for dual-energy CT using a conditional diffusion model

“…The other is that the one-step material decomposition associated with the deep learning-based approaches is much more computationally efficient than the iterative optimization process. Recently, the diffusion model is emerging as a generative approach and gaining much attention in image synthesis [10,11], image translation [12][13][14][15], and image super resolution [16], due to its superior performance to other generative models like the generative adversarial model (GAN) [17]. However, there is still a lack of studies to prove its feasibility in material decomposition applications.…”