Unsupervised Bayesian Generation of Synthetic CT from CBCT Using Patient-Specific Score-Based Prior

TL;DR

This paper introduces an unsupervised, patient-specific diffusion model approach to generate high-quality synthetic CT images from CBCT scans, improving their utility in adaptive radiotherapy by reducing artifacts and HU inaccuracies.

Contribution

It presents a novel unsupervised method using patient-specific score-based priors and TV regularization for CBCT to CT translation, tailored to individual patient anatomy.

Findings

Significant reduction in MAE and increase in PSNR and NCC across multiple cancer sites.

Outperforms two existing diffusion model-based CBCT correction algorithms.

Effective in improving image quality for clinical adaptive radiotherapy applications.

Abstract

Background: Cone-beam computed tomography (CBCT) scans, performed fractionally (e.g., daily or weekly), are widely utilized for patient alignment in the image-guided radiotherapy (IGRT) process, thereby making it a potential imaging modality for the implementation of adaptive radiotherapy (ART) protocols. Nonetheless, significant artifacts and incorrect Hounsfield unit (HU) values hinder their application in quantitative tasks such as target and organ segmentations and dose calculation. Therefore, acquiring CT-quality images from the CBCT scans is essential to implement online ART in clinical settings. Purpose: This work aims to develop an unsupervised learning method using the patient-specific diffusion model for CBCT-based synthetic CT (sCT) generation to improve the image quality of CBCT. Methods: The proposed method is in an unsupervised framework that utilizes a patient-specific score-based model as the image prior alongside a customized total variation (TV) regularization to enforce coherence across different transverse slices. The score-based model is unconditionally trained using the same patient's planning CT (pCT) images to characterize the manifold of CT-quality images and capture the unique anatomical information of the specific patient. The efficacy of the proposed method was assessed on images from anatomical sites including head and neck (H&N) cancer, pancreatic cancer, and lung cancer. The performance of the proposed CBCT correction method was evaluated using quantitative metrics including mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and normalized cross-correlation (NCC). Additionally, the proposed algorithm was benchmarked against two other unsupervised diffusion model-based CBCT correction algorithms.