Adaptive Diffusion Priors for Accelerated MRI Reconstruction

Abstract

Deep MRI reconstruction is commonly performed with conditional models that de-alias undersampled acquisitions to recover images consistent with fully-sampled data. Since conditional models are trained with knowledge of the imaging operator, they can show poor generalization across variable operators. Unconditional models instead learn generative image priors decoupled from the operator to improve reliability against domain shifts related to the imaging operator. Recent diffusion models are particularly promising given their high sample fidelity. Nevertheless, inference with a static image prior can perform suboptimally. Here we propose the first adaptive diffusion prior for MRI reconstruction, AdaDiff, to improve performance and reliability against domain shifts. AdaDiff leverages an efficient diffusion prior trained via adversarial mapping over large reverse diffusion steps. A two-phase reconstruction is executed following training: a rapid-diffusion phase that produces an initial reconstruction with the trained prior, and an adaptation phase that further refines the result by updating the prior to minimize data-consistency loss. Demonstrations on multi-contrast brain MRI clearly indicate that AdaDiff outperforms competing conditional and unconditional methods under domain shifts, and achieves superior or on par within-domain performance.

References in corpus (6)

• Score-Based Generative Modeling through Stochastic Differential Equations
• Inertial Proximal Alternating Linearized Minimization (iPALM) for Nonconvex and Nonsmooth Problems
• Tackling the Generative Learning Trilemma with Denoising Diffusion GANs
• Unsupervised MRI Reconstruction with Generative Adversarial Networks
• One Model to Unite Them All: Personalized Federated Learning of Multi-Contrast MRI Synthesis
• A Few-Shot Learning Approach for Accelerated MRI via Fusion of Data-Driven and Subject-Driven Priors

Cited by in corpus (13)

• High-Frequency Space Diffusion Models for Accelerated MRI
• Data and Physics driven Deep Learning Models for Fast MRI Reconstruction: Fundamentals and Methodologies
• CT Reconstruction using Diffusion Posterior Sampling conditioned on a Nonlinear Measurement Model
• A Tutorial on MRI Reconstruction: From Modern Methods to Clinical Implications
• Self-Supervised Adversarial Diffusion Models for Fast MRI Reconstruction
• Noise Level Adaptive Diffusion Model for Robust Reconstruction of Accelerated MRI
• Fast Inference in Denoising Diffusion Models via MMD Finetuning
• Enhancing quality and speed in database-free neural network reconstructions of undersampled MRI with SCAMPI
• Advancing MRI Reconstruction: A Systematic Review of Deep Learning and Compressed Sensing Integration
• UNISELF: A Unified Network with Instance Normalization and Self-Ensembled Lesion Fusion for Multiple Sclerosis Lesion Segmentation
• Fast MRI Reconstruction Using Deep Learning-based Compressed Sensing: A Systematic Review
• Robust Simultaneous Multislice MRI Reconstruction Using Slice-Wise Learned Generative Diffusion Priors
• MCMC-Correction of Score-Based Diffusion Models for Model Composition