Non-Adversarial Learning: Vector-Quantized Common Latent Space for Multi-Sequence MRI

TL;DR

This paper introduces a non-adversarial, vector-quantized latent space model for multi-sequence MRI translation that improves stability, robustness, and semantic representation over traditional adversarial methods.

Contribution

The paper proposes a novel non-adversarial model using vector-quantized common latent space for multi-sequence MRI synthesis, enhancing stability and semantic understanding.

Findings

Outperforms GAN-based methods on BraTS2021 dataset

Achieves robustness against noise, bias fields, and artifacts

Enables potential one-shot segmentation

Abstract

Adversarial learning helps generative models translate MRI from source to target sequence when lacking paired samples. However, implementing MRI synthesis with adversarial learning in clinical settings is challenging due to training instability and mode collapse. To address this issue, we leverage intermediate sequences to estimate the common latent space among multi-sequence MRI, enabling the reconstruction of distinct sequences from the common latent space. We propose a generative model that compresses discrete representations of each sequence to estimate the Gaussian distribution of vector-quantized common (VQC) latent space between multiple sequences. Moreover, we improve the latent space consistency with contrastive learning and increase model stability by domain augmentation. Experiments using BraTS2021 dataset show that our non-adversarial model outperforms other GAN-based methods, and VQC latent space aids our model to achieve (1) anti-interference ability, which can eliminate the effects of noise, bias fields, and artifacts, and (2) solid semantic representation ability, with the potential of one-shot segmentation. Our code is publicly available.