X-Diffusion: Generating Detailed 3D MRI Volumes From a Single Image Using Cross-Sectional Diffusion Models

2024-04-30

arXiv.org

AI Summary

This study addresses the high cost and prolonged acquisition time of MRI scans by proposing a diffusion-based 2D-to-3D sparse reconstruction method that synthesizes high-fidelity 3D MRI volumes from a single 2D slice. Methodologically, it introduces cross-slice diffusion modeling—treating MRI as an integrated 3D anatomical structure rather than isolated 2D slices—and incorporates a 3D volume-aware training strategy alongside spatial-domain sparse-input modeling. The model demonstrates strong cross-anatomic generalization (e.g., trained on brain MRI, it successfully reconstructs knee MRI). Quantitative evaluation on BraTS and UK Biobank datasets shows significant PSNR improvements over state-of-the-art methods, with accurate recovery of critical anatomical features—including tumor boundaries and spinal curvature. Clinical validation by radiologists confirms its diagnostic credibility and translational potential for accelerating MRI acquisition and reducing patient burden.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Multimodal Learning
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search

Abstract

Magnetic Resonance Imaging (MRI) is a crucial diagnostic tool, but high-resolution scans are often slow and expensive due to extensive data acquisition requirements. Traditional MRI reconstruction methods aim to expedite this process by filling in missing frequency components in the K-space, performing 3D-to-3D reconstructions that demand full 3D scans. In contrast, we introduce X-Diffusion, a novel cross-sectional diffusion model that reconstructs detailed 3D MRI volumes from extremely sparse spatial-domain inputs, achieving 2D-to-3D reconstruction from as little as a single 2D MRI slice or few slices. A key aspect of X-Diffusion is that it models MRI data as holistic 3D volumes during the cross-sectional training and inference, unlike previous learning approaches that treat MRI scans as collections of 2D slices in standard planes (coronal, axial, sagittal). We evaluated X-Diffusion on brain tumor MRIs from the BRATS dataset and full-body MRIs from the UK Biobank dataset. Our results demonstrate that X-Diffusion not only surpasses state-of-the-art methods in quantitative accuracy (PSNR) on unseen data but also preserves critical anatomical features such as tumor profiles, spine curvature, and brain volume. Remarkably, the model generalizes beyond the training domain, successfully reconstructing knee MRIs despite being trained exclusively on brain data. Medical expert evaluations further confirm the clinical relevance and fidelity of the generated images.To our knowledge, X-Diffusion is the first method capable of producing detailed 3D MRIs from highly limited 2D input data, potentially accelerating MRI acquisition and reducing associated costs. The code is available on the project website https://emmanuelleb985.github.io/XDiffusion/ .

Problem

• Generates detailed 3D MRI from single 2D slice.
• Reduces MRI scan time and associated costs.
• Preserves critical anatomical features accurately.

Innovation

• Cross-sectional diffusion model
• 2D-to-3D MRI reconstruction
• Holistic 3D volume modeling