强化集成模型：用于脑MRI修复后处理的SSIM对齐残差优化器
Sharpening the Ensemble: An SSIM-Aligned Residual Refiner for Brain-MRI Inpainting Post-Processing
浏览论文内容
中文总结 AI 辅助
该研究针对脑MRI修复后处理的模糊问题，构建2025年两项并列第一模型的深度集成，训练SSIM对齐的残差优化器，小幅且一致提升BraTS基准的SSIM，无需大规模重训。
中文摘要 AI 辅助
脑MRI修复会将扫描图像中的掩蔽区域替换为合成的、解剖结构合理的健康组织，以便为健康大脑构建的分析工具能够应用于原本会被拒绝的图像。在BraTS局部合成基准测试中，该基准会联合结构相似性指数（SSIM）、峰值信噪比和均方误差（MSE）对提交结果进行排名，目前最强的模型精度较高，但部分模型报告合成区域模糊，且将此归因于训练损失中的ℓ₁和MSE项的均值导向行为。我们在后处理阶段解决该问题，构建2025年两项并列第一模型的深度集成模型，并在集成模型自身输出上训练轻量型残差优化器，训练所用的ℓ₁损失会添加结构相似性项，我们会调整该结构项的权重λ。在适中的λ值下，优化器可提升集成模型的SSIM：在官方评分器的留存复现集上，SSIM从0.8767提升至0.8780；在官方验证排行榜上，SSIM从0.8555提升至0.8572，且MSE基本无变化。该提升幅度虽小但具有一致性，可改善62.6%的留存案例，符号秩检验p值为2.2×10⁻⁷；而过度加重结构项权重则会反转该提升效果。两项 ablation 实验限定了该效果的边界：在两模型集成中加入任意第三个模型都会使其性能下降，经典的非锐化掩蔽在任何强度下都无法提升SSIM（最佳结果为0.8765，低于集成模型的0.8767），因此该提升源于学习到的锐化而非无差别锐化。最终得到一个低成本、可复现的后处理阶段，无需大规模重新训练即可提升已表现强劲的集成模型性能。
英文摘要
Brain-MRI inpainting replaces a masked region of a scan with synthesized, anatomically plausible healthy tissue, so that analysis tools built for healthy brains can be applied to images they would otherwise reject. On the BraTS local-synthesis benchmark, which ranks submissions on the structural similarity index (SSIM), the peak signal-to-noise ratio, and the mean squared error (MSE) jointly, the strongest recent models are accurate, but several report blurry synthesized regions and attribute this to the mean-seeking behavior of the $\ell_1$ and MSE terms in their training losses. We address this in post-processing, forming a deep ensemble of the two co-first-place 2025 models and training a lightweight residual refiner on the ensemble's own outputs under an $\ell_1$ loss augmented with a structural-similarity term whose weight $λ$ we vary. At a moderate $λ$ the refiner improves SSIM over the ensemble, from $0.8767$ to $0.8780$ on a held-out reproduction of the official scorer and from $0.8555$ to $0.8572$ on the official validation leaderboard, with essentially no change in MSE. The gain is small but consistent, improving $62.6\%$ of the held-out cases with a signed-rank $p=2.2\times10^{-7}$, whereas over-weighting the structural term reverses it. Two ablations bound the effect. Adding any third model to the two-model ensemble degrades it, and classical unsharp masking fails to improve SSIM at any strength (best $0.8765$ against $0.8767$), so the gain reflects learned rather than indiscriminate sharpening. The result is a cheap, reproducible post-processing stage that improves an already strong ensemble without any large-scale retraining.
发表机构
- Istanbul Technical University（伊斯坦布尔理工大学）
机构由 AI 辅助整理，请以论文原文为准。
