Is Registering Raw Tagged-MR Enough for Strain Estimation in the Era of Deep Learning?

TL;DR

This paper investigates the effectiveness of deep learning registration methods on raw tagged MRI data for strain estimation, highlighting the impact of tag fading and comparing traditional and modern approaches.

Contribution

It models tag fading considering T1 relaxation and RF pulses, and evaluates how different similarity objectives affect deep learning registration in the context of raw tMRI.

Findings

Traditional similarity metrics struggle with intensity changes due to tag fading.

Deep learning registration methods have limitations when applied directly to raw tMRI.

Harmonic Phase-based methods show robustness to tag fading.

Abstract

Magnetic Resonance Imaging with tagging (tMRI) has long been utilized for quantifying tissue motion and strain during deformation. However, a phenomenon known as tag fading, a gradual decrease in tag visibility over time, often complicates post-processing. The first contribution of this study is to model tag fading by considering the interplay between $T_1$ relaxation and the repeated application of radio frequency (RF) pulses during serial imaging sequences. This is a factor that has been overlooked in prior research on tMRI post-processing. Further, we have observed an emerging trend of utilizing raw tagged MRI within a deep learning-based (DL) registration framework for motion estimation. In this work, we evaluate and analyze the impact of commonly used image similarity objectives in training DL registrations on raw tMRI. This is then compared with the Harmonic Phase-based approach, a traditional approach which is claimed to be robust to tag fading. Our findings, derived from both simulated images and an actual phantom scan, reveal the limitations of various similarity losses in raw tMRI and emphasize caution in registration tasks where image intensity changes over time.