J-Score: Joint Distribution Learning with Score-based Diffusion for Accelerating T1rho Mapping

Congcong Liu, Yuanyuan Liu, Chentao Cao, Jing Cheng, Qingyong Zhu, Tian Zhou, Chen Luo, Yanjie Zhu, Haifeng Wang, Zhuo-Xu Cui, Dong Liang

Abstract: The T1rho mapping technique necessitates acquiring multiple T1rho-weighted images at various spin-lock times (TSL), which results in a lengthy scan time and significantly limits its widespread clinical use. Undersampling is a significant strategy to accelerate T1rho imaging, where it is crucial to model and utilize the joint spatiotemporal correlations priors among different TSL multi-contrast images for high-quality reconstruction. However, current methods that use simplified physical relaxation correlations or black-box deep neural networks to define joint correlations often yield inaccurate results. From a Bayesian framework, the joint distribution provides a powerful capability to represent the joint correlations among multi-contrast T1rho images. Therefore, a new method is introduced to accelerate T1rho parameter imaging, leveraging accurate joint distribution modeling and the captured joint distribution to guide the reconstruction. Specifically, a joint diffusion model is proposed to approximate the joint distribution of multi-contrast T1rho images exploiting the score-matching method. Subsequently, the ill-posed problem caused by the reconstruction of undersampled multi-contrast T1rho images is addressed through the learned joint distribution by employing a constructed joint reverse denoising diffusion model. The superior performance of the proposed method and the capability to accurately characterize the joint distribution was further verified by performing various in vivo experiments. The patient image reconstruction also verifies the feasibility and superiority of the proposed method.

External IDs:doi:10.1109/tmi.2025.3606660