MRI深度学习与影像组学鉴别鼻窦小圆细胞恶性肿瘤
回顾性纳入325例病理确诊的鼻窦恶性肿瘤(SRCMTs 163例,非SRCMTs 162例),基于T1WI、T2WI和CE-T1WI比较深度学习与影像组学模型的术前分类效能。CE-T1WI的ResNet-34平均AUC为0.830,高于CE-T1WI影像组学SVM模型的0.758,但独立测试队列中差异未达统计学显著。
要点
- CE-T1WI的ResNet-34对SRCMTs鉴别效能最高,平均AUC为0.830。
- CE-T1WI深度学习模型敏感性达0.918,NPV为0.879,适合减少漏诊。
- CE-T1WI深度学习与影像组学差异未达统计学显著性,阅片时仍需谨慎解读。
核心结论
- CE-T1WI的ResNet-34网络在SRCMTs与非SRCMTs鉴别中提供较高诊断效能,平均AUC为0.830。
- 本队列中深度学习模型数值上高于MRI影像组学模型,但差异未达到统计学显著性。
研究设计与样本
- 回顾性研究,纳入325例病理确诊鼻窦恶性肿瘤,其中SRCMTs 163例、非SRCMTs 162例。
- 每例均行T1WI、T2WI和CE-T1WI,肿瘤由人工分割。
- 深度学习训练ResNet-18、ResNet-34、ResNet-50、VGG13和VGG16五种卷积神经网络。
- 影像组学每序列提取1200个特征,训练多种机器学习分类器。
- 以AUC、准确性、敏感性、特异性、PPV和NPV评估模型效能,DeLong检验比较AUC。
关键影像表现/征象
- CE-T1WI增强后肿瘤强化特征是区分SRCMTs与非SRCMTs的重要影像依据。
- T1WI与T2WI平扫序列上深度学习与影像组学模型表现大致相当。
诊断标准与关键数值
- CE-T1WI的ResNet-34平均AUC为0.830,准确性0.755,敏感性0.918,特异性0.592,PPV 0.692,NPV 0.879。
- CE-T1WI影像组学SVM模型AUC为0.758,准确性0.755,敏感性0.840,特异性0.667,PPV 0.724,NPV 0.800。
- 独立测试队列中ResNet-34鉴别效能数值上高于影像组学模型,但差异未达统计学显著。
临床应用与阅片要点
- CE-T1WI增强序列是鉴别鼻窦小圆细胞与非小圆细胞恶性肿瘤的关键序列,阅片时应重点关注强化特征。
- 深度学习模型敏感性高而特异性较低,提示可作为筛查工具但阳性预测价值有限。
- T1WI和T2WI上两类模型表现大致相当,平扫序列难以作为独立鉴别依据。
局限性
- 回顾性单队列研究,缺乏多中心外部验证。
- CE-T1WI深度学习与影像组学在独立测试队列中差异未达统计学显著性。
- CE-T1WI的ResNet-34特异性仅0.592,存在较多假阳性。
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
OBJECTIVE: Differentiating sinonasal small round cell malignant tumors (SRCMTs) from non-SRCMTs is challenging due to overlapping MRI features. This study aimed to compare the diagnostic performance of deep learning and radiomics models for preoperative MRI-based classification.
METHODS: We retrospectively analyzed 325 patients with pathologically confirmed sinonasal malignancies (163 SRCMTs and 162 non-SRCMTs). Each patient underwent T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and contrast-enhanced T1-weighted imaging (CE-T1WI). Tumors were manually segmented. Five convolutional neural networks (CNNs)-ResNet-18, ResNet-34, ResNet-50, VGG13, and VGG16-were trained on each sequence. For radiomics, 1200 features were extracted per sequence, and multiple machine learning classifiers were trained. Model performance was assessed by area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The DeLong test was used to compare AUCs between models, with statistical significance set at P < 0.05.
RESULTS: The CE-T1WI-based ResNet-34 model achieved the best performance, with an average AUC of 0.830, the accuracy of 0.755, sensitivity of 0.918, specificity of 0.592, PPV of 0.692, and NPV of 0.879. The corresponding CE-T1WI-based radiomics model using a support vector machine yielded an AUC of 0.758 (accuracy = 0.755, sensitivity = 0.840, specificity = 0.667, PPV = 0.724, NPV = 0.800). On the independent test cohort, ResNet-34 showed numerically higher discriminative performance than the radiomics model, although this difference did not reach statistical significance. For T1WI and T2WI, deep learning and radiomics models demonstrated broadly comparable performance.
CONCLUSIONS: A CE-T1WI-based ResNet-34 network provided high diagnostic efficacy, and in our cohort deep learning models achieved numerically higher comparable performance to MRI-based radiomics models for differentiating SRCMTs from non-SRCMTs.
引用
- Wang, Qu, Li, et al. Deep learning outperformed radiomics based on MRI in the differentiation of sinonasal small round cell and non-small round cell malignant tumors. Eur J Radiol. 2026. PMID: 41604786. doi:10.1016/j.ejrad.2026.112700
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。