让AI更靠谱：用新方法提升脑部MRI分割在不同设备下的准确性与可信度。
Confidence-Calibrating Regularization for Robust Brain MRI Segmentation Under Domain Shift
- 通过特征信息惩罚和置信度错配惩罚，优化模型对域偏移的适应性。
- 在跨扫描仪数据上，分割准确率提升7.4%，不确定性评估误差降低39.5%。
- 仅微调解码器，保持主干网络冻结，适合医疗图像快速部署。
段落分割模型（SAM）在自然图像上表现优异，但在医学影像中存在域偏移和过度自信问题。本文提出轻量级适配框架CalSAM，通过在3D特征图上计算特征费舍尔信息惩罚（FIP）降低编码器对域偏移的敏感性，并通过置信度错配惩罚（CMP）抑制像素级错误的过度自信。联合损失函数\(\mathcal{L}_{\mathrm{CalSAM}}\)仅微调掩码解码器，保持SAM编码器冻结。在跨中心和扫描仪迁移评估中，CalSAM显著提升精度与校准性能：例如在BraTS扫描仪分割任务（西门子→通用电气）中，DSC提升7.4%（80.1% vs. 74.6%），HD95降低26.9%（4.6 mm vs. 6.3 mm），ECE下降39.5%（5.2% vs. 8.6%）。在ATLAS-C（运动伪影）数据集上，DSC提升5.3%（75.9%），ECE降低32.6%（5.8%）。消融实验表明FIP与CMP贡献互补（p<0.01），费舍尔惩罚仅增加约15%训练时间。因此，CalSAM在保留SAM计算优势的同时，提升了脑部MRI分割的域泛化能力与不确定性校准水平。
原文摘要 · Abstract (English)
The Segment Anything Model (SAM) exhibits strong zero-shot performance on natural images but suffers from domain shift and overconfidence when applied to medical volumes. We propose \textbf{CalSAM}, a lightweight adaptation framework that (i) reduces encoder sensitivity to domain shift via a \emph{Feature Fisher Information Penalty} (FIP) computed on 3D feature maps and (ii) penalizes overconfident voxel-wise errors through a \emph{Confidence Misalignment Penalty} (CMP). The combined loss, \(\mathcal{L}_{\mathrm{CalSAM}}\) fine-tunes only the mask decoder while keeping SAM's encoders frozen. On cross-center and scanner-shift evaluations, CalSAM substantially improves accuracy and calibration: e.g., on the BraTS scanner split (Siemens$\to$GE) CalSAM shows a $+7.4\%$ relative improvement in $\mathrm{DSC}$ (80.1\% vs.\ 74.6\%), a $-26.9\%$ reduction in $\mathrm{HD95}$ (4.6 mm vs.\ 6.3 mm), and a $-39.5\%$ reduction in $\mathrm{ECE}$ (5.2\% vs.\ 8.6\%). On ATLAS-C (motion corruptions), CalSAM achieves a $+5.3\%$ relative improvement in $\mathrm{DSC}$ (75.9\%) and a $-32.6\%$ reduction in $\mathrm{ECE}$ (5.8\%). Ablations show FIP and CMP contribute complementary gains ($p<0.01$), and the Fisher penalty incurs a modest $\sim$15\% training-time overhead. CalSAM therefore delivers improved domain generalization and better-calibrated uncertainty estimates for brain MRI segmentation, while retaining the computational benefits of freezing SAM's encoder.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。