基于机器学习的CT影像组学方法用于术前预测进展期胃癌隐匿性腹膜转移
该研究回顾性分析177例进展期胃癌患者,基于动脉期和静脉期CT图像勾画肿瘤核心区及肿瘤肠系膜脂肪间隙共四个感兴趣区,提取1316个影像组学特征并用十种机器学习算法建模,构建影像组学列线图预测隐匿性腹膜转移。肿瘤肠系膜脂肪间隙影像组学在训练集和测试集AUC分别为0.881和0.800(动脉期)、0.838和0.815(静脉期),整合模型训练集AUC为0.943、测试集为0.835,列线图C-index为0.948。
要点
- 肿瘤肠系膜脂肪间隙的影像组学特征与隐匿性腹膜转移相关,动脉期AUC在训练集和测试集分别为0.881和0.800。
- 整合影像组学模型在训练集和测试集的AUC分别为0.943和0.835,均高于各单一影像组学评分。
- 影像组学列线图诊断准确度较好,C-index为0.948,可用于个体化预测隐匿性腹膜转移。
核心结论
- 肿瘤肠系膜脂肪间隙和原发肿瘤的影像组学均与隐匿性腹膜转移相关,CT影像组学列线图对进展期胃癌患者隐匿性腹膜转移具有相对较好的预测性能。
研究设计与样本
- 回顾性分析177例进展期胃癌患者,于CT动脉期和静脉期图像上沿肿瘤最大层面勾画肿瘤核心区及对应肿瘤肠系膜脂肪间隙共四个感兴趣区,每个感兴趣区提取1316个影像组学特征,采用十种机器学习分类算法建模并建立整合影像组学列线图。
诊断标准与关键数值
- 肿瘤肠系膜脂肪间隙影像组学动脉期AUC在训练集为0.881、测试集为0.800,静脉期AUC在训练集为0.838、测试集为0.815。
- 原发肿瘤影像组学动脉期AUC在训练集为0.862、测试集为0.691,静脉期AUC在训练集为0.831、测试集为0.620。
- 整合影像组学模型训练集AUC为0.943,高于各单一影像组学评分的0.831–0.881;测试集AUC为0.835,高于各单一影像组学评分的0.620–0.815。
- 影像组学列线图C-index为0.948。
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
AIM: To develop a machine learning-based CT radiomics model to preoperatively diagnose occult peritoneal metastasis (OPM) in advanced gastric cancer (AGC) patients.
MATERIALS AND METHODS: A total of 177 AGC patients were retrospectively analyzed. Four regions of interest (ROIs) along the largest area of tumor (core ROI) and corresponding tumor mesenteric fat space (peri ROI) were manually delineated on the arterial (A-core and A-peri) and venous phase (V-core and V-peri) of CT images. A total of 1316 radiomics features were extracted from each ROI. Then, ten machine learning classification algorithms were used to develop the model. An integrated radiomics nomogram was established to predict OPM individually.
RESULTS: For the radiomics of tumor mesenteric fat space, the AUCs of A-peri in training and test sets were 0.881 and 0.800, respectively. And the AUCs of V-peri were 0.838 and 0.815, respectively. In terms of primary tumor' s radiomics signature, the AUCs of A-core in training and test sets were 0.862 and 0.691, respectively. The AUCs of V-core were 0.831 and 0.620. Integrated radiomics model showed the highest AUC value when it compared to each single radiomics score in the training (0.943 vs 0.831-0.881) and test set (0.835 vs 0.620-0.815). Radiomics nomogram demonstrated good diagnostic accuracy with a C-index of 0.948.
CONCLUSION: Both the radiomics of tumor mesenteric fat space and primary tumor were associated with OPM. A CT radiomics nomogram had a relatively good predictive performance for detecting OPM in patients with AGC.
引用
- Zhu ZN, Feng QX, Li Q, et al. Machine learning-based CT radiomics approach for predicting occult peritoneal metastasis in advanced gastric cancer preoperatively. Clin Radiol. 2024. PMID: 39571365. doi:10.1016/j.crad.2024.10.008
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。