Cardiac MRI Reconstruction Using Diffusion Priors and Multi-Scale Transformer-Based Feature Modeling.

Consistency Models for Fast MRI Reconstruction Using Regularization by Denoising

Diffusion models (DMs) have emerged as powerful generative priors for MRI reconstruction with promising results. Yet DM-based methods require extensive iterative refinement, limiting their practical deployment. Consistency models (CMs) provide a compelling alternative, aiming to map out the diffusion trajectory in a single pass, enabling faster generation. In this work, we propose CM-RED, a novel MRI reconstruction method that integrates a pretrained CM into the regularization by denoising (RED) scheme. Our method builds on accelerated proximal gradient RED (RED-APG), and further incorporates controlled noise injection during the update steps to enhance generative diversity and accelerate convergence. Extensive experiments on the fastMRI knee and brain datasets demonstrate that CM-RED achieves high-quality reconstructions across multiple anatomies, contrast weights, acceleration factors, and undersampling patterns, using only 4 network function evaluations (NFEs). The proposed method consistently outperforms existing DM- and CM-based approaches in both quantitative metrics and visual fidelity, and exhibits strong robustness to hyperparameter variations, highlighting CM-RED as an efficient and effective generative framework for accelerated MRI reconstruction. The source code and pretrained models are publicly available at https://github.com/MerveGulle/CM-RED.

Multi-Frequency Feature Guided Progressive Divide-and-Conquer Network for Accelerated MRI Reconstruction.

Enhancing Implicit Neural Representations with Image Feature Embedding for Unsupervised Cardiac Cine MRI Reconstruction

This work proposes an image-domain dual-branch INR framework, termed I-FP-INR, which extends the original INR design by introducing an additional feature-processing branch, which aims to extract complementary feature embeddings to enhance the overall representation, thereby benefiting reconstruction.

Output Steering for Latent Diffusion-Based MRI-to-CT Synthesis

Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) are fundamental imaging modalities that provide complementary information for clinical assessment. However, CT acquisition may not always be preferred or available due to additional cost, workflow burden, and exposure to ionizing radiation. Therefore, reliably synthesizing a corresponding CT image from an available MRI scan is an important medical imaging problem. In this work, we present a conditional latent diffusion framework for MRI-to-CT synthesis. To improve the reliability of stochastic generation, the proposed framework incorporates a modular output steering mechanism that favors candidates better matched to target-domain characteristics. In this way, the diversity of diffusion-based synthesis is preserved while output consistency and realism are improved. Experimental results indicate that the proposed approach yields improved quantitative and qualitative performance over a baseline latent diffusion model. The proposed method achieved 26.49±2.19 dB PSNR and 87.89±2.59% SSIM, outperforming the baseline latent diffusion model.

A cross-domain deep learning framework with graph CNN extrapolation for MRI reconstruction

Probabilistic Capsule Diffusion Framework for Uncertainty Aware Medical Image Reconstruction

Reconstruction of medical images is important in improving the diagnosis especially in conditions of noisy and uncertain imaging. Nevertheless, the traditional methods of reconstruction can hardly preserve fine forms of anatomy and do not have effective methods of uncertainty estimation. This paper will introduce a Probabilistic Capsule Diffusion Framework of Uncertainty-Aware Medical Image Reconstruction that combines probabilistic diffusion modeling with capsule-based hierarchical feature learning to enhance the reconstruction accuracy and reliability. The diffusion method allows learning strong probabilistic latent representations, and the capsule network allows the preservation of space and structure. The probabilistic capsule modeling also offers the estimation of uncertainty by variance-based confidence mapping. The proposed framework was tested on the Kaggle Brain MRI data in different noise levels. The results of the experiments showed high performance in terms of Peak Signalto-Noise Ratio of 43.02 dB, Structural Similarity Index of 0.987, and reconstruction accuracy of 99.9% which was much better than the traditional CNN, GAN, and diffusion-based reconstruction algorithms. The framework also had low reconstruction error of 0.008 and high score of confidence of 0.982 which means that it quantifies uncertainty well.