基于冻结自蒸馏特征的CT病灶检测：SALT——空间自适应标签引导温度

中文摘要 AI 辅助

自监督预训练目标具有空间均匀性：图像中教师温度和每个补丁的损失权重处处相同，因此宽度仅为几个补丁的病灶对训练信号的贡献与周围实质组织无差异。现有研究倾向于将视图偏向标注区域，这改变了模型所见内容，但未给目标增加压力。我们转而对自蒸馏的目标进行条件化，该方法名为SALT（空间自适应标签引导温度）。仅在预训练期间可用的弱框衍生标注在编码器的补丁网格上定义了一个紧凑区域，在该区域内，教师的softmax温度被锐化，掩码补丁损失被上调。其他方面的目标、掩码策略和居中统计保持不变，在所有下游使用中，编码器是普通的特征提取器，无标签且无条件化。我们通过冻结编码器并仅训练轻量级多深度CenterNet风格的头部来进行评估，在四个CT队列中进行3D病灶检测，我们将该机制与架构、预训练数据、时间表和标签引导裁剪完全相同但无目标条件化的主干进行隔离验证。我们报告了补丁级可分性、按队列和病灶大小分层的3D检测、框质量，以及无检测器探针——单个冻结补丁嵌入无需配准、掩码或微调即可在随访扫描中重新识别病灶。由于条件化通过空间指标而非标签语义表达，该公式允许任何弱空间标注；我们对病灶进行了实例化和验证。

英文摘要

Self-supervised pretraining objectives are spatially uniform: the teacher temperature and the per-patch loss weight are identical everywhere in the image, so a lesion a few patches wide contributes no more to the training signal than the surrounding parenchyma. Prior work biases the views toward annotated regions, which changes what the model sees but adds no pressure on the objective. We instead condition the targets of self-distillation, a method we call SALT (Spatially Adaptive Label-guided Temperature). Weak, box-derived labels, available only during pretraining, define a compact region on the encoder's patch grid, inside which the teacher's softmax temperature is sharpened and the masked-patch loss is up-weighted. The objectives, the masking policy and the centering statistics are otherwise unchanged, and at every downstream use the encoder is a plain feature extractor with no labels and no conditioning. We evaluate by freezing the encoder and training only a lightweight multi-depth CenterNet-style head, detecting lesions in 3D on four CT cohorts, and we isolate the mechanism against a backbone identical in architecture, pretraining data, schedule and label-guided cropping but with no target conditioning. We report patch-level separability, 3D detection stratified by cohort and by lesion size, box quality, and a detector-free probe in which a single frozen patch embedding re-identifies a lesion in a follow-up scan without registration, masks or fine-tuning. Because the conditioning is expressed through a spatial indicator rather than through label semantics, the formulation admits any weak spatial annotation; we instantiate and validate it for lesions.

发表机构

• University of Kentucky（肯塔基大学）
• University of Louisville（路易斯维尔大学）

机构由 AI 辅助整理，请以论文原文为准。