Department of Diagnostic Radiology, Queen Mary Hospital, Li Ka Shing Faculty of Medicine, University of Hong Kong, Hong Kong, Hong Kong SAR.
Department of Radiology, Pamela Youde Nethersole Eastern Hospital, Hong Kong, Hong Kong SAR.
Eur Radiol. 2021 Jul;31(7):5050-5058. doi: 10.1007/s00330-020-07565-3. Epub 2021 Jan 6.
The study aimed to compare the ability of morphological and texture features derived from contrast-enhanced CT in histological subtyping of epithelial ovarian carcinoma (EOC).
Consecutive 205 patients with newly diagnosed EOC who underwent contrast-enhanced CT were included and dichotomised into high-grade serous carcinoma (HGSC) and non-HGSC. Clinical information including age and cancer antigen 125 (CA-125) was documented. The pre-treatment images were analysed using commercial software, TexRAD, by two independent radiologists. Eight qualitative CT morphological features were evaluated, and 36 CT texture features at 6 spatial scale factors (SSFs) were extracted per patient. Features' reduction was based on kappa score, intra-class correlation coefficient (ICC), univariate ROC analysis and Pearson's correlation test. Texture features with ICC ≥ 0.8 were compared by histological subtypes. Patients were randomly divided into training and testing sets by 8:2. Two random forest classifiers were determined and compared: model 1 incorporating selected morphological and clinical features and model 2 incorporating selected texture and clinical features.
HGSC showed specifically higher texture features than non-HGSC (p < 0.05). Both models performed highly in predicting histological subtypes of EOC (model 1: AUC 0.891 and model 2: AUC 0.937), and no statistical significance was found between the two models (p = 0.464).
CT texture analysis provides objective and quantitative metrics on tumour characteristics with HGSC demonstrating specifically high texture features. The model incorporating texture analysis could classify histology subtypes of EOC with high accuracy and performed as well as morphological features.
• A number of CT morphological and texture features showed good inter- and intra-observer agreements. • High-grade serous ovarian carcinoma showed specifically higher CT texture features than non-high-grade serous ovarian carcinoma. • CT texture analysis could differentiate histological subtypes of epithelial ovarian carcinoma with high accuracy.
本研究旨在比较增强 CT 形态学和纹理特征在卵巢上皮性癌（EOC）组织学亚型中的诊断能力。
连续纳入 205 例新诊断为 EOC 且接受增强 CT 检查的患者，并将其分为高级别浆液性癌（HGSC）和非 HGSC 组。记录了患者的临床资料，包括年龄和癌抗原 125（CA-125）。由两名独立的放射科医生使用商业软件 TexRAD 对术前图像进行分析。评估了 8 种定性 CT 形态学特征，并为每位患者提取 36 种 6 个空间尺度因子（SSF）的 CT 纹理特征。基于 Kappa 评分、组内相关系数（ICC）、单变量 ROC 分析和 Pearson 相关检验进行特征降维。具有 ICC≥0.8 的纹理特征通过组织学亚型进行比较。将患者随机分为训练集和测试集，比例为 8:2。确定并比较了两种随机森林分类器：模型 1 纳入了选择的形态学和临床特征，模型 2 纳入了选择的纹理和临床特征。
HGSC 的纹理特征明显高于非 HGSC（p<0.05）。两种模型在预测 EOC 组织学亚型方面均表现出色（模型 1：AUC 0.891，模型 2：AUC 0.937），且两个模型之间无统计学差异（p=0.464）。
CT 纹理分析为肿瘤特征提供了客观和定量的指标，HGSC 表现出明显更高的纹理特征。纳入纹理分析的模型可以准确地对 EOC 的组织学亚型进行分类，其性能与形态学特征相当。
多项 CT 形态学和纹理特征具有良好的观察者内和观察者间一致性。
高级别浆液性卵巢癌的 CT 纹理特征明显高于非高级别浆液性卵巢癌。
CT 纹理分析可准确区分上皮性卵巢癌的组织学亚型。
