跨癌种肿瘤定位模型，小规模训练实现高精度与泛化能力
A Lightweight Multi-Cancer Tumor Localization Framework for Deployable Digital Pathology
- 基于DenseNet169在4种癌症上训练，用迁移学习实现多癌种定位
- 在4类癌种上达0.97的ROC-AUC，对胰腺癌也达0.71
- 轻量级设计适合部署，可生成兼容现有工具的热力图
从苏木精-伊红染色全幻灯片图像中准确定位肿瘤区域，是空间分析、分子分型和组织结构研究的基础。然而，针对特定癌症训练的深度学习肿瘤检测模型在跨癌种应用时鲁棒性下降。我们探究了在适度规模下跨癌种均衡训练是否可实现高性能并泛化至未见癌种。构建了多癌种肿瘤定位模型MuCTaL，使用DenseNet169在4种癌症（黑色素瘤、肝细胞癌、结直肠癌、非小细胞肺癌）的79,984个非重叠图像块上进行迁移学习训练。该模型在4类训练癌种的验证数据上达到0.97的图像级别ROC-AUC，对独立的胰腺导管腺癌队列也达到0.71。建立可扩展的推理流程，生成与现有数字病理工具兼容的空间肿瘤概率热图。代码与模型已公开于https://github.com/AivaraX-AI/MuCTaL。
原文摘要 · Abstract (English)
Accurate localization of tumor regions from hematoxylin and eosin-stained whole-slide images is fundamental for translational research including spatial analysis, molecular profiling, and tissue architecture investigation. However, deep learning-based tumor detection trained within specific cancers may exhibit reduced robustness when applied across different tumor types. We investigated whether balanced training across cancers at modest scale can achieve high performance and generalize to unseen tumor types. A multi-cancer tumor localization model (MuCTaL) was trained on 79,984 non-overlapping tiles from four cancers (melanoma, hepatocellular carcinoma, colorectal cancer, and non-small cell lung cancer) using transfer learning with DenseNet169. The model achieved a tile-level ROC-AUC of 0.97 in validation data from the four training cancers, and 0.71 on an independent pancreatic ductal adenocarcinoma cohort. A scalable inference workflow was built to generate spatial tumor probability heatmaps compatible with existing digital pathology tools. Code and models are publicly available at https://github.com/AivaraX-AI/MuCTaL.
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。