Coordinate-Based Neural Representation Enabling Zero-Shot Learning for 3D Multiparametric Quantitative MRI

TL;DR

SUMMIT is a novel zero-shot learning framework that enables rapid, accurate, and simultaneous multiparametric 3D MRI reconstruction without external training data, significantly reducing scan times and broadening clinical applications.

Contribution

This work introduces SUMMIT, a physics-informed neural method for zero-shot, multiparametric MRI reconstruction from highly undersampled data, eliminating the need for external training datasets.

Findings

High accuracy in simulated and phantom experiments

Simultaneous reconstruction of multiple MRI parameters

Effective zero-shot learning paradigm for multiparametric imaging

Abstract

Quantitative magnetic resonance imaging (qMRI) offers tissue-specific physical parameters with significant potential for neuroscience research and clinical practice. However, lengthy scan times for 3D multiparametric qMRI acquisition limit its clinical utility. Here, we propose SUMMIT, an innovative imaging methodology that includes data acquisition and an unsupervised reconstruction for simultaneous multiparametric qMRI. SUMMIT first encodes multiple important quantitative properties into highly undersampled k-space. It further leverages implicit neural representation incorporated with a dedicated physics model to reconstruct the desired multiparametric maps without needing external training datasets. SUMMIT delivers co-registered T1, T2, T2*, and quantitative susceptibility mapping. Extensive simulations and phantom imaging demonstrate SUMMIT's high accuracy. Additionally, the proposed unsupervised approach for qMRI reconstruction also introduces a novel zero-shot learning paradigm for multiparametric imaging applicable to various medical imaging modalities.