弱标签下提升细胞病理学多任务预训练，解决低检出率难题
SLAM-AGS: Slide-Label Aware Multi-Task Pretraining Using Adaptive Gradient Surgery in Computational Cytology
- 利用滑片级标签设计双目标优化：负片相似性与正片对比学习
- 在0.5%~10%极低检出率下，袋级F1提升显著，顶400个异常细胞召回率更高
- 自适应梯度手术缓解任务冲突，适合低标注成本的医学图像分析场景
计算细胞学面临两大挑战：实例级标签不可靠且获取成本高昂，样本检出率极低。本文提出SLAM-AGS框架，在滑片级标签指导下联合优化弱监督相似性目标（针对滑片负样本）与自监督对比学习目标（针对滑片正样本），提升下游任务表现。为稳定训练，引入自适应梯度手术以应对任务间梯度冲突，防止模型坍缩。将预训练编码器集成至基于注意力的多实例学习聚合器中，实现袋级预测及异常实例的注意力引导检索。在公开的骨髓细胞学数据集上，模拟0.5%至10%的检出率，结果表明：相较于其他预训练方法，SLAM-AGS在袋级F1-score和前400个阳性细胞检索性能上均有提升，尤其在极低检出率下优势明显，证明梯度干扰缓解可实现更稳定的预训练与更好下游性能。为保障可复现性，完整代码与评估框架已开源：https://github.com/Ace95/SLAM-AGS。
原文摘要 · Abstract (English)
Computational cytology faces two major challenges: i) instance-level labels are unreliable and prohibitively costly to obtain, ii) witness rates are extremely low. We propose SLAM-AGS, a Slide-Label-Aware Multitask pretraining framework that jointly optimizes (i) a weakly supervised similarity objective on slide-negative patches and (ii) a self-supervised contrastive objective on slide-positive patches, yielding stronger performance on downstream tasks. To stabilize learning, we apply Adaptive Gradient Surgery to tackle conflicting task gradients and prevent model collapse. We integrate the pretrained encoder into an attention-based Multiple Instance Learning aggregator for bag-level prediction and attention-guided retrieval of the most abnormal instances in a bag. On a publicly available bone-marrow cytology dataset, with simulated witness rates from 10% down to 0.5%, SLAM-AGS improves bag-level F1-Score and Top 400 positive cell retrieval over other pretraining methods, with the largest gains at low witness rates, showing that resolving gradient interference enables stable pretraining and better performance on downstream tasks. To facilitate reproducibility, we share our complete implementation and evaluation framework as open source: https://github.com/Ace95/SLAM-AGS.
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。