用于多模态IPMN风险分层的堆叠集成的高斯元空间增强
Gaussian Meta-Space Augmentation for Stacking Ensembles in Multimodal IPMN Risk Stratification
- Northwestern University（西北大学）
- Johns Hopkins University（约翰斯·霍普金斯大学）
- Istanbul University（伊斯坦布尔大学）
- Mayo Clinic Florida（佛罗里达州梅奥诊所）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
该研究针对IPMN风险分层问题，提出cUPMI方法并结合多模态信息融合，构建RF堆叠模型，在多中心分析中取得优于基线的性能。
AI中文摘要：
胰腺癌是最致命的恶性肿瘤之一；导管内乳头状黏液瘤（IPMNs）的风险分层为早期干预提供了关键机会，但通常需要侵入性组织活检。主流的基于视觉的方法，包括放射组学和深度学习，提供了有前景但最初相互独立的判别机会。同样，多序列MRI（T1W/T2W）以及胰腺的解剖学分解（头部、体部和尾部）分析提供了额外且潜在互补的信号。这些信息的有效融合对于IPMN异型增生风险的有序预测至关重要，可通过精心正则化和校准的集成堆叠组合器实现。我们提出cUPMI，即对组合器的对数概率元特征进行类条件高斯增强，并在各种预测范式上对其进行测试。在我们的多中心分析中，我们发现cUPMI对适当正则化的L2-逻辑二元分类堆叠的价值有限，但在二元和仅放射组学设置中始终正则化更高容量的树组合器（随机森林（RF）+0.015，XGBoost +0.024的二元AUC，在所有随机种子中均为正）。其最清晰的有序优势出现在XGBoost用于8流放射组学任务（3类：无<低<高，所有种子中QWK提升+0.022）。此外，放射组学与2.5D CNN流的折叠锁定融合产生了最强的整体模型，即RF堆叠，达到QWK 0.595（95%置信区间[0.54, 0.64]）和二元AUC 0.839，优于放射组学、2.5D ResNet和3D DenseNet-121基线模型。
英文摘要：
Pancreatic cancer is among the most lethal malignancies; risk stratification of intraductal papillary mucinous neoplasms (IPMNs) offers a crucial opportunity for early intervention but typically requires invasive tissue biopsy. Dominant vision-based approaches, including radiomics and deep learning, provide promising but initially separate discrimination opportunities. Similarly, multisequence MRI (T1W/T2W) and anatomically decomposed (head, body and tail) analysis of the pancreas provide additional and potentially complementary signals. Effective fusion of this information is crucial in ordinal IPMN dysplasia risk prediction and can be accomplished via a meticulously regularized and calibrated ensemble stacking combiner. We present cUPMI, a class-conditional Gaussian augmentation of a combiner's log-probability meta-features, and test it on various prediction paradigms. In our multi-center analysis, we find cUPMI adds limited value to properly regularized L2-logistic binary classification stacks, but consistently regularizes higher-capacity tree combiners in the binary and radiomics-only setting (RF +0.015 and XGBoost +0.024 binary AUC, positive in all seeds). Its cleanest ordinal benefit appears for XGBoost on an 8-stream radiomics task (3-class no < low < high, +0.022 QWK in all seeds). Separately, fold-locked fusion of radiomics and 2.5D CNN streams yields the strongest overall model, an RF stack reaching QWK 0.595 (95% CI [0.54, 0.64]) and binary AUC 0.839, surpassing radiomics, 2.5D ResNet, and 3D DenseNet-121 baselines.
