A Physics-based Generative Model to Synthesize Training Datasets for MRI-based Fat Quantification

TL;DR

This paper introduces a physics-informed generative model that synthesizes realistic MRI-based fat quantification datasets, enhancing training data availability and improving model robustness across different protocols.

Contribution

The novel Physics-Informed Latent Diffusion Model (PI-LDM) jointly generates quantitative maps and MR images, tailored for q-MRI tasks, incorporating the physical signal generation model.

Findings

Synthetic data achieved comparable quality to state-of-the-art methods (FID: 0.0459).

A U-Net trained with synthetic data showed low bias in fat fraction estimation.

Synthetic datasets improved fat quantification accuracy across different MRI protocols.

Abstract

Deep learning-based techniques have potential to optimize scan and post-processing times required for MRI-based fat quantification, but they are constrained by the lack of large training datasets. Generative models are a promising tool to perform data augmentation by synthesizing realistic datasets. However no previous methods have been specifically designed to generate datasets for quantitative MRI (q-MRI) tasks, where reference quantitative maps and large variability in scanning protocols are usually required. We propose a Physics-Informed Latent Diffusion Model (PI-LDM) to synthesize quantitative parameter maps jointly with customizable MR images by incorporating the signal generation model. We assessed the quality of PI-LDM's synthesized data using metrics such as the Fr\'echet Inception Distance (FID), obtaining comparable scores to state-of-the-art generative methods (FID: 0.0459). We also trained a U-Net for the MRI-based fat quantification task incorporating synthetic datasets. When we used a few real (10 subjects, $~200$ slices) and numerous synthetic samples ($>3000$), fat fraction at specific liver ROIs showed a low bias on data obtained using the same protocol than training data ($0.10\%$ at $\hbox{ROI}_1$, $0.12\%$ at $\hbox{ROI}_2$) and on data acquired with an alternative protocol ($0.14\%$ at $\hbox{ROI}_1$, $0.62\%$ at $\hbox{ROI}_2$). Future work will be to extend PI-LDM to other q-MRI applications.