arXiv 2508.186122025-08-26eess.IVcs.LG

3D nnU-Net用于PET-CT肿瘤分割的跨癌种泛化性压力测试：基于新型食管癌和肺癌数据集的多队列评估

Stress-testing cross-cancer generalizability of 3D nnU-Net for PET-CT tumor segmentation: multi-cohort evaluation with novel oesophageal and lung cancer datasets

QIMR Berghofer Medical Research Institute（昆士兰医学研究 institute）；Australian Institute for Bioengineering and Nanotechnology（澳大利亚生物工程与纳米技术研究所）；The University of Queensland（昆士兰大学）；Frazer Institute（弗拉泽研究所）；Princess Alexandra Hospital（亚历山大公主医院）；Faculty of Health, Medicine and Behavioural Sciences（健康、医学与行为科学学院）；School of Biomedical Sciences（生物医学科学学院）；Uppsala University（乌普萨拉大学）；Rohtak Nuclear Medcare Imaging Therapy and Research Center（罗塔克核医学成像治疗与研究中心）

Soumen Ghosh, Christine Jestin Hannan, Rajat Vashistha, Parveen Kundu, Sandra Brosda, Lauren G. Aoude, James Lonie, Andrew Nathanson, Jessica Ng, Andrew P. Barbour, Viktor Vegh

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本研究通过引入新型食管癌和肺癌数据集，首次对3D nnU-Net在PET-CT肿瘤分割中的跨癌种泛化性进行压力测试，发现联合多队列训练可提升泛化性，数据集多样性是临床鲁棒分割的关键。

摘要：

在解剖部位、扫描仪和患者群体差异巨大的临床PET-CT工作流程中，鲁棒的泛化性对于部署基于深度学习的肿瘤分割至关重要。本研究首次对nnU-Net在PET-CT上进行跨癌种评估，引入两个新型、专家注释的全身数据集：279例食管癌患者（澳大利亚队列）和54例肺癌患者（印度队列）。这些队列补充了公开的AutoPET数据集，可系统地对跨域性能进行压力测试。我们在三种范式下训练和测试3D nnUNet模型：仅目标（食管癌）、仅公开（AutoPET）和联合训练。对于测试集，仅食管癌模型实现了最佳的域内准确率（平均DSC为57.8），但在外部印度肺癌队列上表现极差（平均DSC低于3.4），表明存在严重过拟合；仅公开模型泛化性更广（AutoPET平均DSC为63.5，印度肺癌队列为51.6），但在澳大利亚食管癌队列中表现不佳（平均DSC为26.7）；联合方法提供了最平衡的结果（肺癌平均DSC为52.9，食管癌为40.7，AutoPET为60.9），减少了边界误差并提高了所有队列的鲁棒性。这些发现表明，数据集多样性（尤其是多人口统计学、多中心和多癌种整合）比架构新颖性更能推动鲁棒的泛化性。本研究提出了基于人口统计学的跨癌种深度学习分割评估，并强调数据集多样性而非模型复杂性是临床鲁棒分割的基础。

英文摘要：

Robust generalization is essential for deploying deep learning based tumor segmentation in clinical PET-CT workflows, where anatomical sites, scanners, and patient populations vary widely. This study presents the first cross cancer evaluation of nnU-Net on PET-CT, introducing two novel, expert-annotated whole-body datasets. 279 patients with oesophageal cancer (Australian cohort) and 54 with lung cancer (Indian cohort). These cohorts complement the public AutoPET dataset and enable systematic stress-testing of cross domain performance. We trained and tested 3D nnUNet models under three paradigms. Target only (oesophageal), public only (AutoPET), and combined training. For the tested sets, the oesophageal only model achieved the best in-domain accuracy (mean DSC, 57.8) but failed on external Indian lung cohort (mean DSC less than 3.4), indicating severe overfitting. The public only model generalized more broadly (mean DSC, 63.5 on AutoPET, 51.6 on Indian lung cohort) but underperformed in oesophageal Australian cohort (mean DSC, 26.7). The combined approach provided the most balanced results (mean DSC, lung (52.9), oesophageal (40.7), AutoPET (60.9)), reducing boundary errors and improving robustness across all cohorts. These findings demonstrate that dataset diversity, particularly multi demographic, multi center and multi cancer integration, outweighs architectural novelty as the key driver of robust generalization. This work presents the demography based cross cancer deep learning segmentation evaluation and highlights dataset diversity, rather than model complexity, as the foundation for clinically robust segmentation.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。