使用nnU-Net与后处理的治疗前和治疗后脑转移瘤分割（BraTS 2026）
Pre- and Post-Treatment Brain Metastases Segmentation Using nnU-Net with Post-Processing for BraTS 2026
浏览论文内容
中文总结 AI 辅助
针对BraTS 2026脑转移瘤分割任务，提出基于5折nnU-Net集成与规则后处理的流程，经OOF审计验证稳健性，LW-DSC达0.733/0.751/0.713/0.549，并报告13项阴性结果。
中文摘要 AI 辅助
脑转移瘤在大小、增强模式和治疗后表现方面表现出高度的病灶间变异性，这使得对治疗前和治疗后病例进行体积分割成为BraTS 2026任务1（脑转移瘤）的核心挑战。我们构建了一个基于5折nnU-Net ResEnc-L集成的实用流程，其中每一折独立训练1000个epoch，使用标准的Dice加交叉熵损失，在1296个四模态训练病例上进行。该集成之后是一个基于规则的后处理级联，针对病灶级Dice相似系数（LW-DSC）进行调优，这是一种检测导向的指标，其行为与传统全局Dice非常不同。最终流程在官方验证排行榜上，对增强肿瘤（ET）、肿瘤核心（TC）、全肿瘤（WT）和切除腔（RC）子区域分别达到0.733 / 0.751 / 0.713 / 0.549的LW-DSC。我们不轻信这些排行榜上的提升，而是通过五折袋外（OOF）分析对每个后处理阶段进行审计，该分析在全部1296个训练病例上进行，无模型训练泄漏，并使用官方BraTS评估代码（BraTS_evaluation）评分：结果确认两个阶段是稳健的、逐折一致的改进，而第三个阶段仅提升了排行榜，在袋外未能复现。我们进一步对LW-DSC指标进行了机制分析，解释了为什么召回恢复型后处理风险低，而组件删除则不然，并报告了十三个阴性结果，涵盖损失工程、替代骨干网络和推理时设置，其中几个与广泛持有的直觉相悖。源代码以Apache-2.0许可证发布在https URL。
英文摘要
Brain metastases exhibit high inter-lesion variability in size, enhancement pattern, and post-treatment appearance, making volumetric segmentation of both pre- and post-treatment cases the central challenge of the BraTS 2026 Task 1 (Brain Metastases). We build a pragmatic pipeline on a 5-fold nnU-Net ResEnc-L ensemble, in which each fold is trained independently for 1,000 epochs with the standard Dice + cross-entropy loss on 1,296 four-modality training cases. This ensemble is followed by a rule-based post-processing cascade tuned for the lesion-wise Dice similarity coefficient (LW-DSC), a detection-oriented metric that behaves very differently from the traditional global Dice. The final pipeline reaches an LW-DSC of 0.733 / 0.751 / 0.713 / 0.549 on the enhancing tumour (ET), tumour core (TC), whole tumour (WT), and resection cavity (RC) sub-regions on the official validation leaderboard. Rather than trusting these leaderboard gains, we audit every post-processing stage with a five-fold out-of-fold (OOF) analysis with no model-training leakage over all 1,296 training cases, scored with the official BraTS evaluation code (BraTS_evaluation): it confirms two stages as robust, per-fold-consistent improvements while the third improves only the leaderboard and does not reproduce out-of-fold. We further provide a mechanistic analysis of the LW-DSC metric that explains why recall-recovering post-processing carries low risk whereas component deletion does not, and we report thirteen negative results spanning loss engineering, alternative backbones, and inference-time settings, several of which run counter to widely held intuitions. Source code is released under Apache-2.0 at https://github.com/hornbeamliu/brats2026-met.
发表机构
- Jilin University（吉林大学）
机构由 AI 辅助整理，请以论文原文为准。
