Physics-Driven 3D Gaussian Rendering for Zero-Shot MRI Super-Resolution

TL;DR

This paper introduces a physics-driven, zero-shot 3D Gaussian rendering method for MRI super-resolution that balances data efficiency and computational cost, achieving high-quality reconstructions without extensive training data.

Contribution

It proposes a novel explicit Gaussian representation tailored for MRI, combined with a physics-grounded rendering strategy and parallel computation scheme, enabling efficient zero-shot super-resolution.

Findings

Outperforms existing methods in reconstruction quality

Reduces training and inference time significantly

Requires less annotated data for high-quality results

Abstract

High-resolution Magnetic Resonance Imaging (MRI) is vital for clinical diagnosis but limited by long acquisition times and motion artifacts. Super-resolution (SR) reconstructs low-resolution scans into high-resolution images, yet existing methods are mutually constrained: paired-data methods achieve efficiency only by relying on costly aligned datasets, while implicit neural representation approaches avoid such data needs at the expense of heavy computation. We propose a zero-shot MRI SR framework using explicit Gaussian representation to balance data requirements and efficiency. MRI-tailored Gaussian parameters embed tissue physical properties, reducing learnable parameters while preserving MR signal fidelity. A physics-grounded volume rendering strategy models MRI signal formation via normalized Gaussian aggregation. Additionally, a brick-based order-independent rasterization scheme enables highly parallel 3D computation, lowering training and inference costs. Experiments on two public MRI datasets show superior reconstruction quality and efficiency, demonstrating the method's potential for clinical MRI SR.