-
目前,胃癌是最常见的恶性肿瘤之一,也是全球第二大癌症相关死亡原因[1],大约70%的病例发生在亚洲,仅中国就占了一半以上[2]。淋巴结转移是胃癌预后因素之一,在选择合适的新辅助化疗可行方案中起着关键的作用[3-4]。传统影像学检查确定淋巴结转移的方式主要是基于淋巴结的大小。然而,反应性或炎症性淋巴结可以增大、正常或者轻微增大,因此部分病人存在临床淋巴结分期不准确的风险[6]。计算机断层成像是术前评估淋巴结状态最常用的成像手段,但是据目前报道来看准确率尚不能令人满意,仅60%。能谱CT在降低扫描剂量的同时, 能获得高质量的影像图像,而纹理分析通过高通量特征提取,分析并挖掘图像中的定量特征,从而客观评估病灶的均质性[7]。本研究旨在探讨以能谱CT为基础的纹理分析在胃癌术前淋巴结转移中的评估价值。
-
进行能谱扫描的80位胃癌术前病人,其中发生淋巴结转移的病人有49例(其中男40例,女9例),年龄39~80岁, 而未发生淋巴结转移的病人有31例(其中男24例,女7例),年龄46~86岁。在所选影像组学特征中,398个经Mann-Whitney U检验评估具有统计学意义(P < 0.05)。经单因素logistic分析筛选出10个特征具有统计学意义(P < 0.05),然后经过MRMR方法选择要素子集,保留了10个相关纹理特征,这10个纹理特征的性能见表 1,在经过多元logistics分析及似然比检验选择后,最终保留了4个纹理特征,其纹理特征性能见表 2,并构建了预测模型,最终模型显示出了良好的预测性能,AUC值约为0.79(0.69~0.89)。将选取的4个纹理特征构建预测模型,用组学评分计算公式表示:Rad-score=0.577 107 400 004 795×(Intercept)-2.053 355 886 945 71×Correlation_angle45_offset1-0.695 288 789 630 599×Correlation_angle0_offset4+1.209 241 722 435 82×Inertia_AllDirection_offset1_SD-1.208 985 369 906 08×Inertia_angle90_offset1。
特征名称 准确度 准确度下限 准确度上限 敏感度 特异度 45°相关偏离量1 0.68 0.56 0.78 0.71 0.61 90°惯性偏离量1 0.61 0.50 0.72 0.57 0.68 90°相关偏离量7 0.68 0.56 0.78 0.73 0.58 全方位惯性偏离量1-SD 0.59 0.47 0.70 0.39 0.90 90°相关偏离量4 0.64 0.52 0.74 0.61 0.68 全方位相关偏离量1 0.66 0.55 0.76 0.71 0.58 90°逆差距偏离量1 0.66 0.55 0.76 0.80 0.45 熵差 0.60 0.48 0.71 0.41 0.90 45°惯性偏离量1 0.65 0.54 0.75 0.63 0.68 0°相关偏离量4 0.58 0.46 0.68 0.33 0.97 表 1 降维后筛选的10个纹理特征的性能
变量单位概率比 95%CI P 45°相关偏离量1 0.13[0.03~0.52] 0.004 175 0°相关偏离量4 0.50[0.23~1.11] 0.087 074 全方位惯性偏离量1-SD 3.35[0.91~12.34] 0.069 100 90°惯性偏离量1 0.30[0.06~1.51] 0.144 051 表 2 经过多元logistics分析及似然比检验选择后最终保留4个纹理特征的置信区间(CI)及P值
经过多重交叉验证得出准确性、灵敏度及特异性,训练组准确性、灵敏度及特异性分别为0.79、0.82及0.77;实验组准确性、灵敏度及特异性分别为0.69、0.69及0.68。
基于能谱CT纹理分析在预测胃癌术前淋巴结转移中的价值
Value of texture analysis based on spectral CT in predicting preoperative lymph node metastasis of gastric cancer
-
摘要:
目的探讨能谱CT纹理分析在胃癌病人术前预测淋巴结转移中的价值。 方法回顾性分析80例经手术切除病理证实的胃癌病人(训练组为57例,验证组为23例)。使用专用的纹理分析软件AK对术前静脉期能谱70 keV CT图像进行分割病变并提取影像学特征。使用Mann-Whitney U检验分析2组之间的特征,保留P < 0.05的特征,进一步通过logistic回归分析寻找有鉴别能力的特征(P < 0.05),然后使用最小冗余最大相关方法剔除冗余,但保留与标签相关性最高的10个特征,使用逐步多元logistic回归,构建预测模型,并构建最终模型,通过ROC分析对模型的性能进行评价。 结果纹理特征方面,最小冗余最大相关方法选择的10个放射组学相关特征对训练组及验证组的鉴别能力较好(AUC>0.64);多元logistic回归预测模型的AUC为0.79(0.69~0.89)。 结论基于能谱CT的纹理分析有望成为胃癌病人术前淋巴结转移预测的非侵入性工具。 -
关键词:
- 胃肿瘤 /
- 淋巴结转移 /
- 纹理分析 /
- 体层摄影术,X线计算机
Abstract:ObjectiveTo study the value of spectral CT texture analysis in predicting preoperative lymph node metastasis in patients with gastric cancer. MethodsEighty patients(including 57 cases in training group and 23 cases in verification group) with gastric cancer confirmed by surgical resection and pathology were retrospectively analyzed.The special texture analysis software AK was used to segment the lesions and extract the imaging features on the preoperative spectral CT 70 keV venous phase cross-sectional images.The Mann-Whitney U test was used to analyze the features between two groups, and the P < 0.05 features were preserved.The discriminative features were further found by single factor logistic regression analysis.The minimum redundancy maximum correlation method(MRMR) was used to eliminate the 10 features with the highest correlation with the label.Stepwise multiple logistic regression was used to construct the prediction model and final model.The performance of the model was evaluated using ROC analysis. ResultsIn terms of texture features, the 10 radiology-related features selected by MRMR had better discriminative ability for training group and verification group(AUC>0.64), and the AUC of multivariate logistic regression prediction model was 0.79(0.69-0.89). ConclusionsThe texture analysis based on spectral CT is expected to be a non-invasive tool for predicting preoperative lymph node metastasis in patients with gastric cancer. -
Key words:
- gastric neoplasms /
- lymph node metastasis /
- texture analysis /
- tomography, X-ray computer
-
表 1 降维后筛选的10个纹理特征的性能
特征名称 准确度 准确度下限 准确度上限 敏感度 特异度 45°相关偏离量1 0.68 0.56 0.78 0.71 0.61 90°惯性偏离量1 0.61 0.50 0.72 0.57 0.68 90°相关偏离量7 0.68 0.56 0.78 0.73 0.58 全方位惯性偏离量1-SD 0.59 0.47 0.70 0.39 0.90 90°相关偏离量4 0.64 0.52 0.74 0.61 0.68 全方位相关偏离量1 0.66 0.55 0.76 0.71 0.58 90°逆差距偏离量1 0.66 0.55 0.76 0.80 0.45 熵差 0.60 0.48 0.71 0.41 0.90 45°惯性偏离量1 0.65 0.54 0.75 0.63 0.68 0°相关偏离量4 0.58 0.46 0.68 0.33 0.97 表 2 经过多元logistics分析及似然比检验选择后最终保留4个纹理特征的置信区间(CI)及P值
变量单位概率比 95%CI P 45°相关偏离量1 0.13[0.03~0.52] 0.004 175 0°相关偏离量4 0.50[0.23~1.11] 0.087 074 全方位惯性偏离量1-SD 3.35[0.91~12.34] 0.069 100 90°惯性偏离量1 0.30[0.06~1.51] 0.144 051 -
[1] BRAY F, FERLAY J, SOERJOMATARAM I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2018, 68(6): 394. doi: 10.3322/caac.21492 [2] RUSSO AE, STRONG VE. Gastric cancer etiology and management in asia and the west[J]. Annu Rev Med, 2019, 70: 353. doi: 10.1146/annurev-med-081117-043436 [3] CHEN W, ZHENG R, BAADE PD, et al. Cancer statistics in China, 2015[J]. CA Cancer J Clin, 2016, 66(2): 115. doi: 10.3322/caac.21338 [4] FUKAGAWA T, KATAI H, MIZUSAWA J, et al. A prospective multi-institutional validity study to evaluate the accuracy of clinical diagnosis of pathological stage Ⅲ gastric cancer (JCOG1302A)[J]. Gastric Cancer, 2018, 21(1): 68. doi: 10.1007/s10120-017-0701-1 [5] SMYTH EC, VERHEIJ M, ALLUM W, et al. Gastric cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up[J]. Ann Oncol, 2016, 27(Suppl 5): v38. [6] WU S, ZHENG J, LI Y, et al. Development and validation of an mri-based radiomics signature for the preoperative prediction of lymph node metastasis in bladder cancer[J]. EBio Medicine, 2018, 34: 76. [7] LUBNER MG, SMITH AD, SANDRASEGARAN K, et al. CT texture analysis: definitions, applications, biologic correlates, and challenges[J]. Radiographics, 2017, 37(5): 1483. doi: 10.1148/rg.2017170056 [8] KANESAKA T, NAGAHAMA T, UEDO N, et al. Clinical predictors of histologic type of gastric cancer[J]. Gastrointest Endosc, 2018, 87(4): 1014. doi: 10.1016/j.gie.2017.10.037 [9] 徐成, 胡月珍, 张再军, 等. CT能谱扫描及40 keV对应CT值等参数对肺内良、恶性肿块的诊疗价值分析[J]. 实用癌症杂志, 2019, 34(1): 89. [10] LIU JY, DENG JY, ZHANG NN, et al. Clinical significance of skip lymph-node metastasis in pN1 gastric-cancer patients after curative surgery[J]. Gastroenterol Rep (Oxf), 2019, 7(3): 193. doi: 10.1093/gastro/goz008 [11] 柴亚如, 高剑波, 邢静静, 等. 能谱CT定量参数对胃癌淋巴的定性评估价值[J]. 中华胃肠外科杂志, 2017, 20(3): 309. doi: 10.3760/cma.j.issn.1671-0274.2017.03.016 [12] FEHRENBACH U, FELDHAUS F, KAHN J, et al. Tumour response in non-small-cell lung cancer patients treated with chemoradiotherapy-Can spectral CT predict recurrence?[J]. J Med Imaging Radiat Oncol, 2019, 63(5): 641. doi: 10.1111/1754-9485.12926 [13] LIU Z, WANG S, DONG D, et al. The applications of radiomics in precision diagnosis and treatment of oncology: opportunities and challenges[J]. Theranostics, 2019, 9(5): 1303. doi: 10.7150/thno.30309 [14] LAMBIN P, LEIJENAAR RTH, DEIST TM, et al. Radiomics: the bridge between medical imaging and personalized medicine[J]. Nat Rev Clin Oncol, 2017, 14(12): 749. doi: 10.1038/nrclinonc.2017.141 [15] MA Z, FANG M, HUANG Y, et al. CT-based radiomics signature for differentiating Borrmann type Ⅳ gastric cancer from primary gastric lymphoma[J]. Eur J Radiol, 2017, 91: 142. doi: 10.1016/j.ejrad.2017.04.007 [16] LIU S, LIU S, JI C, et al. Application of CT texture analysis in predicting histopathological characteristics of gastric cancers[J]. Eur Radiol, 2017, 27(12): 4951. doi: 10.1007/s00330-017-4881-1 [17] XU Y, LU L, E LN, et al. Application of radiomics in predicting the malig-nancy of pulmonary nodules in different sizes[J]. AJR Am J Roentgenol, 2019, 213(6): 1213. doi: 10.2214/AJR.19.21490 [18] ZHANG R, XU L, WEN X, et al. A nomogram based on bi-regional radiomics features from multimodal magnetic resonance imaging for preoperative prediction of microvascular invasion in hepatocellular carcinoma[J]. Quant Imaging Med Surg, 2019, 9(9): 1503. doi: 10.21037/qims.2019.09.07 [19] LI Y, ERESEN A, LU Y, et al. Radiomics signature for the preoperative assessment of stage in advanced colon cancer[J]. Am J Cancer Res, 2019, 9(7): 1429. [20] WANG Y, LIU W, YU Y, et al. CT radiomics nomogram for the preoperative prediction of lymph node metastasis in gastric cancer[J]. Eur Radiol, 2020, 30(2): 976. doi: 10.1007/s00330-019-06398-z [21] MA Z, LIANG C, HUANG Y, et al. Can lymphovascular invasion be predicted by preoperative multiphasic dynamic CT in patients with advanced gastric cancer?[J]. Eur Radiol, 2017, 27(8): 3383. doi: 10.1007/s00330-016-4695-6 [22] LI J, DONG D, FANG M, et al. Dual-energy CT-based deep learning radiomics can improve lymph node metastasis risk prediction for gastric cancer[J]. Eur Radiol, 2020, 30(4): 2324. doi: 10.1007/s00330-019-06621-x [23] 陆中元, 陈兵, 刘淼, 等. CT能谱曲线对非小细胞肺癌胸内淋巴结转移诊断价值[J]. 临床军医杂志, 2016, 44(2): 200.