Keep the Core: Adversarial Priors for Significance-Preserving Brain MRI Segmentation

TL;DR

This paper introduces a novel data augmentation framework for brain MRI segmentation that preserves critical diagnostic features by identifying and selectively augmenting or masking important regions using adversarial priors.

Contribution

The paper proposes SAGE and KEEP modules that leverage adversarial optimization to identify essential features and guide augmentation, improving segmentation robustness and generalization.

Findings

Achieves state-of-the-art robustness on 2D medical datasets.

Effectively preserves diagnostic features during augmentation.

Enhances model generalization without inference overhead.

Abstract

Medical image segmentation is constrained by sparse pathological annotations. Existing augmentation strategies, from conventional transforms to random masking for self-supervision, are feature-agnostic: they often corrupt critical diagnostic semantics or fail to prioritize essential features. We introduce "Keep the Core," a novel data-centric paradigm that uses adversarial priors to guide both augmentation and masking in a significance-preserving manner. Our approach uses SAGE (Sparse Adversarial Gated Estimator), an offline module identifying minimal tokens whose micro-perturbation flips segmentation boundaries. SAGE forges the Token Importance Map $W$ by solving an adversarial optimization problem to maximally degrade performance, while an $\ell_1$ sparsity penalty encourages a compact set of sensitive tokens. The online KEEP (Key-region Enhancement \& Preservation) module uses $W$ for a two-pronged augmentation strategy: (1) Semantic-Preserving Augmentation: High-importance tokens are augmented, but their original pixel values are strictly restored. (2) Guided-Masking Augmentation: Low-importance tokens are selectively masked for an $\text{MAE}$-style reconstruction, forcing the model to learn robust representations from preserved critical features. "Keep the Core" is backbone-agnostic with no inference overhead. Extensive experiments show SAGE's structured priors and KEEP's region-selective mechanism are highly complementary, achieving state-of-the-art segmentation robustness and generalization on 2D medical datasets.