纹理分析作为鉴别肾脏肿瘤的影像组学标志物
回顾性纳入119例经病理证实肾脏肿瘤且行腹部CT检查的患者,对肿瘤手动分割并提取纹理特征,另用支持向量机分类。直方图偏度与峰度鉴别透明细胞亚型与嗜酸细胞瘤AUC为0.91和0.93,中位数鉴别乳头状亚型与嗜酸细胞瘤AUC为0.99;机器学习进一步将透明细胞亚型与其他肿瘤、乳头状亚型与其他肿瘤的鉴别AUC提升至0.91和0.92。作者认为纹理分析是有前景的无创工具。
要点
- 直方图偏度与峰度鉴别透明细胞亚型与嗜酸细胞瘤AUC分别为0.91和0.93。
- 直方图中位数鉴别乳头状亚型与嗜酸细胞瘤AUC为0.99,鉴别嗜酸细胞瘤与其他肿瘤AUC为0.92。
- 机器学习鉴别透明细胞亚型与其他肿瘤、乳头状亚型与其他肿瘤AUC分别为0.91和0.92。
核心结论
- 纹理分析在CT图像上鉴别肾脏肿瘤是有前景的无创工具,应用机器学习后结果进一步改善,支持将纹理分析发展为鉴别各类肾脏肿瘤的定量生物标志物。
研究设计与样本
- 经IRB批准的回顾性分析,纳入119例病理证实的肾脏肿瘤且有腹部CT检查的患者,对肿瘤CT图像手动分割后行纹理分析,并应用支持向量机方法对肿瘤类型进行分类。
诊断标准与关键数值
- 直方图特征偏度鉴别透明细胞亚型与嗜酸细胞瘤AUC为0.91(p<0.0001)。
- 直方图特征峰度鉴别透明细胞亚型与嗜酸细胞瘤AUC为0.93(p<0.0001)。
- 直方图特征中位数鉴别乳头状亚型与嗜酸细胞瘤AUC为0.99(p<0.0001),鉴别嗜酸细胞瘤与其他肿瘤AUC为0.92。
- 机器学习鉴别透明细胞亚型与其他肿瘤AUC为0.91,鉴别乳头状亚型与其他肿瘤AUC为0.92。
临床应用与阅片要点
- 作者认为纹理分析是有前景的无创工具,可用于在CT图像上区分肾脏肿瘤。
- 作者认为结果在应用机器学习后进一步改善,支持进一步将纹理分析发展为鉴别各类肾脏肿瘤的定量生物标志物。
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
PURPOSE: To evaluate the utility of texture analysis for the differentiation of renal tumors, including the various renal cell carcinoma subtypes and oncocytoma.
MATERIALS AND METHODS: Following IRB approval, a retrospective analysis was performed, including all patients with pathology-proven renal tumors and an abdominal computed tomography (CT) examination. CT images of the tumors were manually segmented, and texture analysis of the segmented tumors was performed. A support vector machine (SVM) method was also applied to classify tumor types. Texture analysis results were compared to the various tumors and areas under the curve (AUC) were calculated. Similar calculations were performed with the SVM data.
RESULTS: One hundred nineteen patients were included. Excellent discriminators of tumors were identified among the histogram-based features noting features skewness and kurtosis, which demonstrated AUCs of 0.91 and 0.93 (p < 0.0001), respectively, for differentiating clear cell subtype from oncocytoma. Histogram feature median demonstrated an AUC of 0.99 (p < 0.0001) for differentiating papillary subtype from oncocytoma and an AUC of 0.92 for differentiating oncocytoma from other tumors. Machine learning further improved the results achieving very good to excellent discrimination of tumor subtypes. The ability of machine learning to distinguish clear cell subtype from other tumors and papillary subtype from other tumors was excellent with AUCs of 0.91 and 0.92, respectively.
CONCLUSION: Texture analysis is a promising non-invasive tool for distinguishing renal tumors on CT images. These results were further improved upon application of machine learning, and support the further development of texture analysis as a quantitative biomarker for distinguishing various renal tumors.
引用
- Yu H, Scalera J, Khalid M, et al. Texture analysis as a radiomic marker for differentiating renal tumors. Abdom Radiol (NY). 2017. PMID: 28421244. doi:10.1007/s00261-017-1144-1
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。