整合[18F]FDG PET/CT影像组学与临床特征的可解释机器学习模型预测胃肠道淋巴瘤化疗后穿孔:多中心研究
本研究纳入两家医院257例胃肠道淋巴瘤患者,提取FDG高摄取区域的PET/CT影像组学特征,结合临床特征用五种机器学习算法建模预测化疗后穿孔。临床-PET/CT联合模型(逻辑回归)预测性能优于单一模型,训练、内部验证和外部验证队列AUC分别为0.918、0.858和0.852,SHAP分析显示影像组学特征贡献显著。
要点
- 临床-PET/CT联合模型预测化疗后穿孔性能最佳,外部验证AUC为0.852
- C反应蛋白升高和T细胞非霍奇金淋巴瘤是预测穿孔的临床危险因素
- SHAP分析显示PET/CT影像组学特征对穿孔预测有显著贡献
核心结论
- 整合[18F]FDG PET/CT影像组学特征与临床特征的机器学习联合模型可为胃肠道淋巴瘤化疗后穿孔风险评估提供临床可用的列线图。
研究设计与样本
- 多中心研究纳入两家医院共257例胃肠道淋巴瘤患者,收集临床数据与PET/CT图像,从FDG高摄取区域提取影像组学特征,用五种机器学习算法构建临床、PET/CT影像组学及临床-PET/CT联合模型,并用SHAP方法实现模型可解释性。
主要发现
- 训练队列144例(平均年龄56.90±14.69岁,88例男性)中22例诊断为化疗后穿孔,内部验证队列63例(平均年龄61.76±13.71岁,36例男性)中11例,外部验证队列50例(平均年龄63.28±12.17岁,27例男性)中11例。
- 临床特征中,C反应蛋白水平升高和T细胞非霍奇金淋巴瘤病理类型是预测化疗后穿孔的危险因素。
- SHAP分析显示PET/CT影像组学特征对化疗后穿孔的预测有显著贡献。
- 采用逻辑回归算法的临床-PET/CT联合模型预测性能优于采用ExtraTrees算法的单一临床模型和PET/CT影像组学模型,训练队列AUC分别为0.918、0.872和0.857,内部验证队列分别为0.858、0.795和0.849,外部验证队列分别为0.852、0.770和0.807。
关键数值与推荐
- 临床-PET/CT联合模型(逻辑回归)在训练、内部验证和外部验证队列的AUC分别为0.918、0.858和0.852。
- 临床模型(ExtraTrees)在训练、内部验证和外部验证队列的AUC分别为0.872、0.795和0.770。
- PET/CT影像组学模型(ExtraTrees)在训练、内部验证和外部验证队列的AUC分别为0.857、0.849和0.807。
临床怎么用
- 对于胃肠道淋巴瘤患者,化疗前可基于[18F]FDG PET/CT影像组学特征联合临床特征(如C反应蛋白和病理类型)评估化疗后穿孔风险,辅助优化治疗决策。
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)。
原文
BACKGROUND: Perforation following chemotherapy in gastrointestinal lymphoma (PFCGL) is a rare but severe and life-threatening complication. Early pre-chemotherapy prediction is crucial for optimizing treatment and improving outcomes, yet it remains challenging. This study aimed to develop a machine learning (ML) model combining clinical features and [18F]FDG PET/CT radiomics signature to predict the risk of PFCGL. MATERIALS AND METHODS: A total of 257 patients diagnosed at two hospitals with gastrointestinal lymphoma were included. Clinical data and PET/CT images were collected. PET/CT radiomics signature were extracted from the FDG-avid regions. Five ML algorithms were used to develop predictive models, including clinical, PET/CT radiomics, and clinical-PET/CT models. The SHapley Additive exPlanations (SHAP) method was applied for model interpretability. RESULTS: The internal training, internal validation and external validation cohorts consisted of 144 patients (mean age: 56.90 ± 14.69 years; 88 males), 63 patients (mean age: 61.76 ± 13.71 years; 36 males) and 50 patients (mean age: 63.28 ± 12.17 years; 27 males), respectively. Among these, 22, 11, and 11 patients were diagnosed with PFCGL in the three cohorts, correspondingly. For clinical features, elevated C-reactive protein levels and pathological type of T-cell non-Hodgkin's lymphoma were risk factors for predicting PFCGL. SHAP analysis revealed that PET/CT radiomics signature significantly contributed to the prediction of PFCGL. The clinical-PET/CT model utilizing the logistic regression (LR) algorithm exhibited superior predictive performance for PFCGL compared to the individual clinical model and the PET/CT radiomics model employing the ExtraTrees algorithm. This was evidenced by the area under the curve (AUC) values of 0.918, 0.872 and 0.857 in the training cohort, 0.858, 0.795 and 0.849 in the internal validation cohort, and 0.852, 0.770 and 0.807 in the external validation cohort, respectively. CONCLUSION: The combined model that integrates [18F]FDG PET/CT radiomics signature with clinical features through the application of machine learning provides a clinically actionable nomogram for the evaluation and prediction the risk of PFCGL.
引用
- Chen D, Li T, Cui J, et al. An interpretable machine learning model integrating [18F]FDG PET/CT radiomics and clinical features for predicting perforation following chemotherapy in gastrointestinal lymphoma: a multicenter study. Eur J Nucl Med Mol Imaging. 2026. PMID: 42151619. doi:10.1007/s00259-026-07921-3
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。