Single-subject Multi-contrast MRI Super-resolution via Implicit Neural Representations

TL;DR

This paper introduces a subject-specific super-resolution method for multi-contrast MRI using Implicit Neural Representations, enabling high-quality 3D reconstructions from anisotropic 2D views with minimal training time.

Contribution

It presents a novel INR-based framework that jointly models multi-contrast MRI views for super-resolution, which is trained quickly on a single GPU and improves 3D image quality.

Findings

Achieves realistic super-resolution across contrast pairs.

Converges to optimal mutual information, ensuring anatomical fidelity.

Trains within minutes on commodity hardware.

Abstract

Clinical routine and retrospective cohorts commonly include multi-parametric Magnetic Resonance Imaging; however, they are mostly acquired in different anisotropic 2D views due to signal-to-noise-ratio and scan-time constraints. Thus acquired views suffer from poor out-of-plane resolution and affect downstream volumetric image analysis that typically requires isotropic 3D scans. Combining different views of multi-contrast scans into high-resolution isotropic 3D scans is challenging due to the lack of a large training cohort, which calls for a subject-specific framework. This work proposes a novel solution to this problem leveraging Implicit Neural Representations (INR). Our proposed INR jointly learns two different contrasts of complementary views in a continuous spatial function and benefits from exchanging anatomical information between them. Trained within minutes on a single commodity GPU, our model provides realistic super-resolution across different pairs of contrasts in our experiments with three datasets. Using Mutual Information (MI) as a metric, we find that our model converges to an optimum MI amongst sequences, achieving anatomically faithful reconstruction. Code is available at: https://github.com/jqmcginnis/multi_contrast_inr/