放射组学与基础模型融合用于可解释的肾细胞癌分类：内部基准测试与探索性外部迁移
Radiomics--Foundation Fusion for Interpretable RCC Classification: Internal Benchmarking and Exploratory External Transfer
浏览论文内容
中文总结 AI 辅助
本研究在KiTS23上比较放射组学与基础模型特征及其融合用于肾细胞癌亚型分类，发现3D MedVAE门控融合最佳（内部AUC 82.7%），且放射组学仍为可解释的互补成分。
中文摘要 AI 辅助
从增强CT中对肾细胞癌（RCC）进行准确的术前亚型分类在临床上仍具挑战性，因为透明细胞肾细胞癌（ccRCC）和非透明细胞肾细胞癌常表现出重叠的影像表现。本研究评估基础模型表示是否减少了对人工设计的放射组学特征的依赖，或者放射组学在可解释的肿瘤表征中是否仍具有互补作用。我们在KiTS23数据集上比较了放射组学、传统CNN特征、MedicalNet预训练特征、MedVAE表示及其融合变体在二元ccRCC分类中的性能，报告了受试者工作特征曲线下面积（AUC）及其自助法置信区间，并报告平均精度（AP）作为对类别不平衡敏感的补充指标。我们进一步评估了分支移除消融实验、TCGA/AIMI外部迁移，以及使用放射组学置换重要性和门控级别分析的可解释性。在内部测试中，3D MedVAE门控融合取得了最佳性能，AUC为82.7%，AP为92.2%。在外部TCGA队列上，同一模型实现了AUC为79.5%和AP为98.9%，但由于仅有2例外部非ccRCC病例，特异性仍不确定。门控分析显示以放射组学为主导的融合机制，表明基础模型表示充当了基于病例的细化信号，而非结构化肿瘤描述符的替代品。这些发现支持放射组学作为基础模型时代基于CT的RCC表征中互补且临床可解释的组成部分。
英文摘要
Accurate preoperative subtype classification of renal cell carcinoma (RCC) from contrast-enhanced CT remains clinically challenging because clear cell RCC (ccRCC) and non-clear cell RCC often show overlapping imaging appearances. This study evaluates whether foundation representations reduce reliance on handcrafted radiomics, or whether radiomics remains complementary for interpretable tumour characterisation. We compared radiomics, conventional CNN features, MedicalNet-pretrained features, MedVAE representations, and fusion variants for binary ccRCC classification on KiTS23, reporting area under the receiver operating characteristic curve (AUC) with bootstrap confidence intervals and average precision (AP) as a complementary class-imbalance-sensitive metric. We further assessed branch-removal ablation, TCGA/AIMI external transfer, and interpretability using radiomics permutation importance and gate-level analysis. Internally, 3D MedVAE gated fusion achieved the best performance, with an AUC of 82.7% and AP of 92.2%. On the external TCGA cohort, the same model achieved an AUC of 79.5% and AP of 98.9%, although specificity remains uncertain because only two external non-ccRCC cases were available. Gate analysis showed a radiomics-dominant fusion regime, suggesting that foundation representations acted as case-dependent refinement signals rather than replacements for structured tumour descriptors. These findings support radiomics as a complementary and clinically interpretable component of CT-based RCC characterisation in the foundation-model era.
发表机构
- Research Ireland Centre for Research Training in Machine Learning（爱尔兰研究机构机器学习研究培训中心）
- University College Dublin（都柏林大学学院）
- School of Veterinary Medicine, University College Dublin（都柏林大学学院兽医学院）
机构由 AI 辅助整理，请以论文原文为准。
