Search papers, labs, and topics across Lattice.
This paper introduces the Pixel-Level Residual Diffusion Transformer (PRDiT), a novel framework designed to generate high-resolution 3D CT volumes with fine details while addressing the computational and optimization challenges of existing generative models. By employing a two-stage training architecture that combines a local denoiser with a global residual diffusion transformer, PRDiT effectively separates low-frequency structures and refines high-frequency residuals, resulting in improved training stability and optimization. Experimental results on the LIDC-IDRI and RAD-ChestCT datasets show that PRDiT significantly outperforms state-of-the-art models in terms of 3D FID, MMD, and Wasserstein distance scores.
PRDiT achieves unprecedented detail in 3D CT volume generation while simplifying the optimization process, outperforming leading models in the field.
Generating high-resolution 3D CT volumes with fine details remains challenging due to substantial computational demands and optimization difficulties inherent to existing generative models. In this paper, we propose the Pixel-Level Residual Diffusion Transformer (PRDiT), a scalable generative framework that synthesizes high-quality 3D medical volumes directly at voxel-level. PRDiT introduces a two-stage training architecture comprising 1) a local denoiser in the form of an MLP-based blind estimator operating on overlapping 3D patches to separate low-frequency structures efficiently, and 2) a global residual diffusion transformer employing memory-efficient attention to model and refine high-frequency residuals across entire volumes. This coarse-to-fine modeling strategy simplifies optimization, enhances training stability, and effectively preserves subtle structures without the limitations of an autoencoder bottleneck. Extensive experiments conducted on the LIDC-IDRI and RAD-ChestCT datasets demonstrate that PRDiT consistently outperforms state-of-the-art models, such as HA-GAN, 3D LDM and WDM-3D, achieving significantly lower 3D FID, MMD and Wasserstein distance scores.
