预训练、课程调优与集成：面向AutoPET V的示踪剂感知交互式分割流程

中文摘要 AI 辅助

全身PET/CT中的病变交互式分割要求模型既能提供可靠的初始预测，又能在推理过程中对稀疏的修正涂鸦做出高效响应。该场景极具挑战性，因为FDG与PSMA研究中的示踪剂分布、生理摄取模式、病变外观及采集特征存在显著差异。我们提出TRIAGE，即示踪剂感知的解剖引导交互式分割优化（Tracer-aware Refinement via Interactive Anatomy-Guided sEgmentation）。核心骨干为3D STU-Net，通过带异步掩码策略的掩码自编码预训练进行初始化，旨在学习可迁移的解剖及跨模态表征，再进行任务特定微调。我们同步训练辅助器官分割模型，其预测结果提供明确的解剖上下文，助力区分生理摄取与恶性病变。专用示踪剂分类器将每项研究路由至FDG或PSMA特定分支；各分支内，第一阶段分割模型利用CT、PET及器官上下文生成初始病变掩码，初始预测再结合累积的前景/背景涂鸦，由第二个交互式分割网络进行细化。FDG与PSMA分支共享相同整体处理流程，但独立训练以适配示踪剂特定外观与误差模式。我们还采用课程式训练与模型集成，提升交互步骤及异质性队列的鲁棒性。实验基于官方AutoPET V数据及十倍划分开展；定量结果、消融实验及最终测试集性能留作占位符，待挑战赛评估后补充。代码：this https URL。

英文摘要

Interactive lesion segmentation in whole-body PET/CT requires a model to provide a strong initial prediction while also responding efficiently to sparse corrective scribbles during inference. This setting is particularly challenging because tracer distributions, physiological uptake patterns, lesion appearance, and acquisition characteristics differ substantially between FDG and PSMA studies. We present TRIAGE, Tracer-aware Refinement via Interactive Anatomy-Guided sEgmentation. The core backbone is a 3D STU-Net initialized through masked autoencoding pre-training with an asynchronous masking strategy, aiming to learn transferable anatomical and cross-modal representations before task-specific fine-tuning. In parallel, we train an auxiliary organ segmentation model whose predictions provide explicit anatomical context and help distinguish physiological uptake from malignant lesions. A dedicated tracer classifier first routes each study to an FDG- or PSMA-specific branch. Within each branch, a first-stage segmentation model consumes CT, PET, and organ context to generate an initial lesion mask. The initial prediction is then combined with cumulative foreground/background scribbles and refined by a second interactive segmentation network. The FDG and PSMA branches share the same overall processing pipeline but are trained independently to account for tracer-specific appearance and error modes. We additionally employ curriculum-style training and model ensembling to improve robustness across interaction steps and heterogeneous cohorts. Experiments are conducted using the official AutoPET V data and ten-fold split; quantitative results, ablations, and final test-set performance are left as placeholders to be completed after the challenge evaluation. Code: https://github.com/Liiiii2101/AUTOPET2026-MEDAI.

发表机构

• The Netherlands Cancer Institute（荷兰癌症研究所）
• Radboud University Medical Centre（拉德堡德大学医学中心）
• Macao Polytechnic University（澳门理工大学）
• Amsterdam University Medical Center（阿姆斯特丹大学医学中心）

机构由 AI 辅助整理，请以论文原文为准。