Manifold Model for High-Resolution fMRI Joint Reconstruction and Dynamic Quantification

TL;DR

This paper introduces OSSIMM, a physics-based manifold model for high-resolution fMRI that improves reconstruction quality and enables dynamic quantification of parameters like R2* maps with high temporal resolution.

Contribution

The paper presents a novel manifold model tailored for OSSI fMRI data, enhancing reconstruction and enabling dynamic parameter quantification, outperforming traditional subspace models.

Findings

Achieves 12x acceleration in high-resolution fMRI reconstruction.

Outperforms subspace models in reconstruction quality.

Enables dynamic R2* mapping at 150 ms temporal resolution.

Abstract

Oscillating Steady-State Imaging (OSSI) is a recent fMRI acquisition method that exploits a large and oscillating signal, and can provide high SNR fMRI. However, the oscillatory nature of the signal leads to an increased number of acquisitions. To improve temporal resolution and accurately model the nonlinearity of OSSI signals, we build the MR physics for OSSI signal generation as a regularizer for the undersampled reconstruction rather than using subspace models that are not well suited for the data. Our proposed physics-based manifold model turns the disadvantages of OSSI acquisition into advantages and enables joint reconstruction and quantification. OSSI manifold model (OSSIMM) outperforms subspace models and reconstructs high-resolution fMRI images with a factor of 12 acceleration and without spatial or temporal resolution smoothing. Furthermore, OSSIMM can dynamically quantify important physics parameters, including $R_2^*$ maps, with a temporal resolution of 150 ms.