基于CT的可解释机器学习与深度学习模型鉴别嗜铬细胞瘤与肾上腺皮质腺瘤：多中心研究
本研究回顾性纳入两中心共401例患者，从双期CT提取1781个影像组学特征，比较八种机器学习分类器与SqueezeNet深度学习模型。基于逻辑回归的整合模型表现最佳，内部测试集AUC为0.917、准确率86.4%，外部验证AUC为0.909；SHAP分析显示依赖非均匀性为关键诊断驱动因素，并构建了临床列线图。
要点
- LR整合模型在内外验证中AUC均约0.91，肿瘤≥4 cm时AUC达0.933，诊断稳定性高。
- SHAP提示依赖非均匀性是鉴别PHEO与ACA的关键影像组学特征。
- 基于SHAP的列线图可提供个体化、透明的风险估计，辅助术前决策。
核心结论
- SHAP解释的整合LR模型是鉴别PHEO与ACA的稳健无创工具，结合影像组学与临床数据支持精准手术规划和个体化管理。
研究设计与样本
- 回顾性多中心研究，纳入两中心401例患者；中心1（n=331）分为训练集232例、验证集49例、内部测试集50例，中心2（n=70）为外部验证队列。
- 从双期CT提取1781个影像组学特征，比较八种机器学习分类器与基于SqueezeNet的深度学习模型，以AUC和决策曲线分析量化性能。
关键影像表现/征象
- SHAP分析识别依赖非均匀性为关键诊断驱动因素。
诊断标准与关键数值
- LR整合模型内部测试集AUC为0.917（95%CI 0.820-0.990），准确率86.4%；外部验证AUC为0.909（95%CI 0.7995-1.000）。
- LR整合模型整体AUC为0.850（95%CI 0.838-0.869），优于其他分类器及DL基线。
- 亚组分析中肿瘤≥4 cm的AUC为0.933（95%CI 0.821-1.000）。
临床应用与阅片要点
- 可借助整合模型与列线图对肾上腺偶发瘤进行术前无创风险评估，辅助鉴别PHEO与ACA。
- 关注依赖非均匀性这一影像组学特征，其被SHAP识别为关键诊断驱动因素。
- 肿瘤≥4 cm时模型诊断稳定性高，可优先参考模型输出辅助手术规划。
局限性
- 研究为回顾性设计，外部验证样本量较小（n=70）。
- 深度学习仅采用SqueezeNet单一架构作为基线。
以上解读依据论文摘要（全文未开放获取或许可不允许翻译转载）。
英文原文（摘要）· 点击展开
BACKGROUND: Differentiating pheochromocytoma (PHEO) from adrenocortical adenoma (ACA) is vital to avoid intraoperative hypertensive crises or redundant surgeries.
PURPOSE: To identify the optimal diagnostic strategy by benchmarking various machine learning (ML) and deep learning (DL) architectures for the preoperative differentiation of PHEO from ACA.
METHODS: We retrospectively enrolled 401 patients from two centers. Center 1 (n = 331) was divided into training (n = 232), validation (n = 49), and internal test (n = 50) sets; Center 2 (n = 70) served as an external validation cohort. From dual-phase CT, 1,781 radiomics features were extracted. We benchmarked eight ML classifiers against a SqueezeNet-based DL model. Performance was quantified using AUC and decision curve analysis. Shapley Additive Explanations (SHAP) were applied to interpret the best-performing model and construct a clinical nomogram.
RESULTS: The Logistic Regression (LR)-based integrated model outperformed other classifiers and the DL baseline (AUC: 0.850, 95% CI: 0.838-0.869)), achieving an AUC of 0.917 (95% CI: 0.820-0.990) with an accuracy of 86.4% in the internal test set and 0.909 (95% CI: 0.7995-1.000) in external validation. Subgroup analysis showed high diagnostic stability, with an AUC of 0.933 (95% CI: 0.821-1.000) for tumors ≥ 4 cm. SHAP analysis identified dependence non-uniformity as a key diagnostic driver. The resulting nomogram provided personalized, transparent risk estimation.
CONCLUSION: The SHAP-interpreted integrated LR model is a robust, non-invasive tool for distinguishing PHEO from ACA. By combining radiomic and clinical data, this transparent framework supports precise surgical planning and individualized management.
引用
- Sun H, Yu M, Zhou Y, et al. Interpretable machine learning and deep learning model for discriminating pheochromocytoma from adrenocortical adenoma based on CT: A multicenter study. Eur J Radiol. 2026. PMID: 42107147. doi:10.1016/j.ejrad.2026.112902
供执业医师工作参考，不替代临床判断；数值与分级以指南原文为准。