Transformer-based Parameter Fitting of Models derived from Bloch-McConnell Equations for CEST MRI Analysis

TL;DR

This paper introduces a transformer-based neural network for fitting parameters in CEST MRI data, significantly improving upon classical methods by leveraging self-supervised learning to analyze complex physiological signals.

Contribution

The novel application of a transformer neural network for parameter fitting in CEST MRI, trained self-supervised, enhances accuracy over traditional gradient-based solvers.

Findings

Transformer model outperforms classical solvers in accuracy.

Self-supervised training improves model robustness.

Applicable to in-vitro CEST spectra analysis.

Abstract

Chemical exchange saturation transfer (CEST) MRI is a non-invasive imaging modality for detecting metabolites. It offers higher resolution and sensitivity compared to conventional magnetic resonance spectroscopy (MRS). However, quantification of CEST data is challenging because the measured signal results from a complex interplay of many physiological variables. Here, we introduce a transformer-based neural network to fit parameters such as metabolite concentrations, exchange and relaxation rates of a physical model derived from Bloch-McConnell equations to in-vitro CEST spectra. We show that our self-supervised trained neural network clearly outperforms the solution of classical gradient-based solver.