构建可解释的多模态影像模型鉴别T1期鼻咽癌与良性增生
回顾性纳入2015至2022年经组织学确诊的161例患者(T1-NPC 95例、BH 66例),基于MRI和内镜提取特征构建影像组学及MRI-内镜联合模型,MLP联合模型测试集平均AUC达0.98、准确度0.90、精确度0.90、敏感度0.93、特异度0.86,SHAP分析显示纹理特征(GLSZM、GLCM、GLRLM)和一阶特征对鉴别至关重要。
要点
- MRI-内镜联合模型鉴别T1-NPC与BH的测试集平均AUC为0.98,优于单纯MRI影像组学模型。
- MLP联合模型测试集敏感度0.93、特异度0.86、准确度0.90、精确度0.90。
- SHAP显示纹理特征(GLSZM、GLCM、GLRLM)和一阶特征为鉴别关键特征。
核心结论
- 与传统影像组学方法相比,结合MRI和内镜的多模态模型更准确地鉴别良恶性组织;SHAP可可视化特征贡献和模型预测,突显模型的临床潜力。
研究设计与样本
- 回顾性收集2015至2022年经组织学确诊的161例患者数据,包括T1-NPC 95例和BH 66例,基于MRI和内镜勾画感兴趣区提取特征。
诊断标准与关键数值
- MLP联合模型测试集平均AUC为0.98,准确度0.90,精确度0.90,敏感度0.93,特异度0.86。
- SHAP分析显示纹理特征(包括GLSZM、GLCM和GLRLM)和一阶特征对区分T1-NPC与BH至关重要。
临床应用与阅片要点
- MRI-内镜联合模型在测试集平均AUC达0.98,可辅助鉴别T1-NPC与BH。
- SHAP可可视化特征贡献和模型预测,有助于理解模型判断依据。
局限性
- 原文未报告局限性。
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 AND OBJECTIVE: Differentiating T1-stage nasopharyngeal carcinoma (NPC) from benign hyperplasia (BH) is challenging. This study aims to construct and validate a multimodal model combining magnetic resonance imaging (MRI) and endoscopy to distinguish T1-NPC from BH. Additionally, SHapley Additive exPlanations (SHAP) are used for model interpretability analysis.
MATERIALS AND METHODS: Data from 161 patients with histologically confirmed diagnoses between 2015 and 2022 were retrospectively collected, including 95 cases of T1-NPC and 66 cases of BH. Regions of interest (ROI) were drawn based on MRI and endoscopy to extract features. Feature selection techniques, such as elastic net, recursive feature elimination, and deep learning, were used to identify the optimal feature subset. Naive Bayes, Adaptive Boosting (AdaBoost), Light Gradient Boosting Machine (LightGBM), k-Nearest Neighbors (kNN), and Multilayer Perceptron (MLP) were applied to establish the MRI radiomics model and the MRI-endoscopy combined model. SHAP was used to perform interpretability analysis of the models.
RESULTS: The MRI-endoscopy combined model outperformed the radiomics model, with the MLP-based model showing the best performance. The mean AUC of the test set reached 0.98, with an accuracy of 0.90, precision of 0.90, sensitivity of 0.93, and specificity of 0.86. SHAP analysis revealed that texture features (including GLSZM, GLCM, and GLRLM) and first-order features were critical for distinguishing T1-NPC from BH.
CONCLUSION: Compared to traditional radiomics methods, the multimodal model combining MRI and endoscopy more accurately distinguishes between benign and malignant tissues. SHAP enables visualization of feature contributions and model predictions, highlighting the model's clinical potential.
引用
- Wang KJ, Lin TC, Zuo SJ, et al. Construction of an interpretable multimodal image model for differentiating T1-stage nasopharyngeal carcinoma from benign hyperplasia. Eur J Radiol. 2026. PMID: 41966588. doi:10.1016/j.ejrad.2026.112814
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。