为免疫荧光显微成像跨数据集评估构建鲁棒视觉编码器
Building Robust Vision Encoders for Cross-Dataset Evaluation in Immunofluorescent Microscopy
把细胞图像通道按「上下文/概念」分组，用上下文通道作参考、以掩码知识蒸馏训练通道自适应编码器，使模型在未见通道配置的 IF 数据集上免重训直接迁移——这套「通道条件化表征 + 蒸馏」思路可迁移到多模态/多通道输入不一致的视觉表征学习场景。
原文依据 C3R is a two-fold framework comprising a channel-adaptive encoder architecture and a masked knowledge distillation training strategy
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论免疫荧光（IF）图像能够揭示亚细胞层面结构与功能的详细信息。然而，与 RGB 图像不同，由于不同实验室与研究采用不同的染色方案，IF 数据集在通道数量与通道配置上存在不一致，这给深度学习模型带来了挑战。尽管现有方法构建了用于训练的通道自适应模型，但它们并未在具有未见通道配置的 IF 数据集之间开展评估。为解决这一问题，我们首先引入一种具有生物学依据的细胞图像通道视角，将通道划分为上下文（context）与概念（concept）两类，并把图像中的上下文通道视为概念通道的参考。基于这一视角，我们提出通道条件化细胞表征（Channel Conditioned Cell Representations，C3R）框架，该框架学习到的表征能够良好迁移到分布内（ID）数据集与分布外（OOD）数据集，二者分别具有相同与不同的通道配置。C3R 是一个双组成部分的框架，包含通道自适应编码器架构与掩码知识蒸馏训练策略，二者均围绕上下文—概念原则构建。我们发现，C3R 在 ID 与 OOD 任务上均优于现有基准，并在 CHAMMI 基准的冻结编码器评估上取得最优结果。我们的方法为 IF 数据集之间的跨数据集泛化开辟了新路径，且无需在未见通道配置上重新训练。
英文原文摘要 1505 字符
Immunofluorescence (IF) images reveal detailed information about structures and functions at the subcellular level. However, unlike RGB images, IF datasets pose challenges for deep learning models due to their inconsistencies in channel count and configuration, stemming from varying staining protocols across laboratories and studies. Although existing approaches build channel-adaptive models for training, they do not perform evaluations across IF datasets with unseen channel configurations. To address this, we first introduce a biologically informed view of cellular image channels by grouping them into either context or concept, where we treat the context channels as a reference for the concept channels in the image. We leverage this view to propose Channel Conditioned Cell Representations (C3R), a framework that learns representations that transfers well to both in-distribution (ID) and out-of-distribution (OOD) datasets which contain same and different channel configurations, respectively. C3R is a two-fold framework comprising a channel-adaptive encoder architecture and a masked knowledge distillation training strategy, both built around the context-concept principle. We find that C3R outperforms existing benchmarks on both ID and OOD tasks, while yielding state-of-the-art results on frozen encoder evaluation on the CHAMMI benchmark. Our method opens a new pathway for cross-dataset generalization between IF datasets, with no need for retraining on unseen channel configurations.