Data-Efficient Meningioma Segmentation via Implicit Spatiotemporal Mixing and Sim2Real Semantic Injection

TL;DR

This work tackles data-efficiency in meningioma segmentation by decoupling anatomical background diversity from lesion semantics. It introduces a dual-augmentation framework: (i) spatiotemporal mixing using Implicit Neural Representations to produce continuous, topology-preserving diffeomorphic deformations, and (ii) Sim2Real semantic injection that transplant real lesion textures into healthy backgrounds via a distance-field guided fusion. The approach yields significant gains across diverse models, with notable improvements in Dice scores and boundary precision, and demonstrates that synthetic data can closely approximate real distributions while providing complementary diversity. The framework offers a practical, data-efficient pathway for robust medical image analysis under annotation constraints, with potential impact on clinical AI deployment in low-resource settings.

Abstract

The performance of medical image segmentation is increasingly defined by the efficiency of data utilization rather than merely the volume of raw data. Accurate segmentation, particularly for complex pathologies like meningiomas, demands that models fully exploit the latent information within limited high-quality annotations. To maximize the value of existing datasets, we propose a novel dual-augmentation framework that synergistically integrates spatial manifold expansion and semantic object injection. Specifically, we leverage Implicit Neural Representations (INR) to model continuous velocity fields. Unlike previous methods, we perform linear mixing on the integrated deformation fields, enabling the efficient generation of anatomically plausible variations by interpolating within the deformation space. This approach allows for the extensive exploration of structural diversity from a small set of anchors. Furthermore, we introduce a Sim2Real lesion injection module. This module constructs a high-fidelity simulation domain by transplanting lesion textures into healthy anatomical backgrounds, effectively bridging the gap between synthetic augmentation and real-world pathology. Comprehensive experiments on a hybrid dataset demonstrate that our framework significantly enhances the data efficiency and robustness of state-of-the-art models, including nnU-Net and U-Mamba, offering a potent strategy for high-performance medical image analysis with limited annotation budgets.

Figures (3)

• Figure 1: Overview of the proposed Dual-Augmentation Framework. It includes the INR-based spatial augmentation pathway, the semantic injection-based synthesis pathway, and the final weighted sampling training mechanism.
• Figure 2: Qualitative demonstration of data augmentation effects. Figure 2(A) illustrates the Semantic Augmentation, which fills gaps in the pathological distribution by realistically embedding real lesion textures into healthy backgrounds. Figure 2(B) displays Spatial Augmentation based on INR velocity field mixing.
• Figure 3: Visualization comparison of ablation studies. Each row represents a typical test sample. The green contour represents the Ground Truth (GT) annotated by experts, and the red contour represents the prediction results of each variant model.