M2Diff: Multi-Modality Multi-Task Enhanced Diffusion Model for MRI-Guided Low-Dose PET Enhancement

TL;DR

M2Diff is a novel multi-modality multi-task diffusion model that enhances low-dose PET scans by separately processing MRI and PET data to better leverage modality-specific features, improving reconstruction quality.

Contribution

The paper introduces M2Diff, a multi-modality multi-task diffusion model that processes MRI and PET separately and fuses features hierarchically for improved low-dose PET reconstruction.

Findings

Outperforms existing methods on brain datasets.

Achieves superior qualitative and quantitative results.

Effective in both healthy and Alzheimer's disease cases.

Abstract

Positron emission tomography (PET) scans expose patients to radiation, which can be mitigated by reducing the dose, albeit at the cost of diminished quality. This makes low-dose (LD) PET recovery an active research area. Previous studies have focused on standard-dose (SD) PET recovery from LD PET scans and/or multi-modal scans, e.g., PET/CT or PET/MRI, using deep learning. While these studies incorporate multi-modal information through conditioning in a single-task model, such approaches may limit the capacity to extract modality-specific features, potentially leading to early feature dilution. Although recent studies have begun incorporating pathology-rich data, challenges remain in effectively leveraging multi-modality inputs for reconstructing diverse features, particularly in heterogeneous patient populations. To address these limitations, we introduce a multi-modality multi-task diffusion model (M2Diff) that processes MRI and LD PET scans separately to learn modality-specific features and fuse them via hierarchical feature fusion to reconstruct SD PET. This design enables effective integration of complementary structural and functional information, leading to improved reconstruction fidelity. We have validated the effectiveness of our model on both healthy and Alzheimer's disease brain datasets. The M2Diff achieves superior qualitative and quantitative performance on both datasets.