INR能走多远？基于跨域参数高效的INR脑MRI语义分割
How Far Can INRs Go? Cross-Domain Parameter-efficient INR-Based Semantic Segmentation for Brain MRI
浏览论文内容
中文总结 AI 辅助
本研究分析跨域脑MRI分割中INR的性能机制，发现其在低参数下优势显著，并提出层次化架构HierINRSeg，较MetaSeg在域内外Dice分别提升5.6和8.2个百分点。
中文摘要 AI 辅助
生物医学图像分割是医学图像分析的核心，但实际部署常面临标注有限、内存约束和跨站点分布偏移等挑战。隐式神经表示（INRs）最近作为一种轻量级语义分割替代方案出现，以远少于传统架构的参数实现了具有竞争力的性能。然而，基于INR的分割机制、扩展行为及域泛化能力仍未被充分理解。在本工作中，我们针对跨域脑MRI分割研究这些问题。我们在低参数设置下分析基于INR的分割，并在域内和域外设置中与传统流程进行比较。令人惊讶的是，我们发现基于INR的模型并非简单地随参数预算增加而改善。其优势在低参数和有限增强设置下最为显著，而基于U-Net的模型则更多受益于更大容量和标准增强。我们还研究了INR如何在隐藏特征中编码语义信息，并表明互补的分割相关结构分布在多个INR层中。基于这一见解，我们引入了HierINRSeg，一种层次化基于INR的架构，通过聚合多层表示以提高鲁棒性和泛化能力。大量实验表明，HierINRSeg始终优于MetaSeg（一种近期强大的基于INR的分割基线），在域内测试集上Dice平均提升5.6个百分点，在域外提升8.2个百分点。总体而言，我们的分析确定了基于INR的分割最有效的条件，为模型选择和未来研究提供了具体指导。
英文摘要
Biomedical image segmentation is central to medical image analysis, but practical deployment often faces limited annotations, memory constraints, and cross-site distribution shifts. Implicit Neural Representations (INRs) have recently emerged as a lightweight alternative for semantic segmentation, achieving competitive performance with substantially fewer parameters than conventional architectures. However, the mechanisms, scaling behavior, and domain generalization abilities of INR-based segmentation remain insufficiently understood. In this work, we study these questions in the context of cross-domain brain MRI segmentation. We analyze INR-based segmentation across low-parameter regimes, comparing it with conventional pipelines in both in-domain and out-of-domain settings. Surprisingly, we find that INR-based models do not simply improve with increasing parameter budget. Their advantage is most pronounced under low-parameter and limited-augmentation settings, while U-Net-based models benefit more from larger capacity and standard augmentation. We also investigate how INRs encode semantic information in their hidden features and show that complementary segmentation-relevant structure is distributed across multiple INR layers. Building on this insight, we introduce HierINRSeg, a hierarchical INR-based architecture that aggregates multi-layer representations for improved robustness and generalization. Extensive experiments show that HierINRSeg consistently outperforms MetaSeg, a strong recent INR-based segmentation baseline, with an average improvement of 5.6 percentage points in Dice for the in-domain test set and 8.2 percentage points out-of-domain. Overall, our analysis identifies the conditions under which INR-based segmentation is most effective, providing concrete guidance for model selection and future research.
发表机构
- University of Waterloo（滑铁卢大学）
机构由 AI 辅助整理，请以论文原文为准。
