论著摘要 |【CT】从治疗前CT纹理的部位间肿瘤异质性的新型代表特征通过临床结果分类卵巢癌(双语版)

2017-08-27 10:22:53 admin 52

A novel representation of inter-site tumour heterogeneity from pre-treatment computedtomography textures classifies ovarian cancers by clinical outcome.

发表日期:2017.4.9    来源:Eur Radiol. 

作者:Vargas HA1Veeraraghavan H2Micco M3Nougaret S3,4,5Lakhman Y3Meier AA3Sosa R3Soslow RA6Levine DA7Weigelt B6Aghajanian C8Hricak H3Deasy J2Snyder A8Sala E3.

作者介绍

    1.Department of Radiology, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY, 10065, USA. 

    2.Department of Medical Physics, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY, 10065, USA.

    3.Department of Radiology, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY, 10065, USA.

    4.Service de Radiologie, Institut Régional du Cancer de Montpellier, Montpellier, France.

    5.INSERM, U1194, Institut de Recherche en Cancérologie de Montpellier (IRCM), Montpellier, France.

    6.Department of Pathology, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY, 10065, USA.

    7.Department of Surgery, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY, 10065, USA.

    8.Department of Medicine, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY, 10065, USA.

目的

CT评估高级浆液性卵巢癌(HGSOC)患者临床结局与影像组学得到的部位间空间异质性指标在多发性转移性病变的之间的关系。

To evaluate the associations between clinical outcomes and radiomics-derived inter-site spatial heterogeneity metrics across multiple metastatic lesions on CT in patients with high-grade serous ovarian cancer (HGSOC).

方法

IRB批准的38HGSOC患者的回顾性研究。手动分割术前CT怀疑HGSOC的所有部位。从每个肿瘤部位计算基于灰度级相关矩阵的纹理,并使用高斯混合模型将其分成五组。计算成对的部位间相似性,生成部位间相似矩阵(ISM)。从ISM计算出部位间纹理异质性指标,并与临床结果进行比较。


IRB-approved retrospective study of 38 HGSOC patients. All sites of suspected HGSOC involvement on preoperative CT were manually segmented. Gray-level correlation matrix-based textures were computed from each tumour site, and grouped into five clusters using a Gaussian Mixture Model. Pairwise inter-site similarities were computed, generating an inter-site similarity matrix (ISM). Inter-site texture heterogeneity metrics were computed from the ISM and compared to clinical outcomes.

结果

在评估的12个位点间纹理异质性指标中,捕获的部位间纹理相似度差异与较短的总体生存期(站点间相似熵,相似性群集阴影和部位间相似性群集突出;p0.05 )和不完全的手术切除(相似度集群阴影,部位间相似度集群突出和部位间聚类方差;p0.05)相关联。每个患者的疾病部位总数和每个患者的总肿瘤体积与总生存率均不相关。涉及细胞周期蛋白E1基因(CCNE1)的19q12的扩增主要发生在部位间纹理异质性更高的患者中。



Of the 12 inter-site texture heterogeneity metrics evaluated, those capturing the differences in texture similarities across sites were associated with shorter overall survival (inter-site similarity entropy, similarity level cluster shade, and inter-site similarity level cluster prominence; p ≤ 0.05) and incomplete surgical resection (similarity level cluster shade, inter-site similarity level cluster prominence and inter-site cluster variance; p ≤ 0.05). Neither the total number of disease sites per patient nor the overall tumour volume per patient was associated with overall survival. Amplification of 19q12 involving cyclin E1 gene (CCNE1) predominantly occurred in patients with more heterogeneous inter-site textures.

结论

非侵入性捕获空间的部位间异质性的定量指标可预测HGSOC患者的结局。



Quantitative metrics non-invasively capturing spatial inter-site heterogeneity may predict outcomes in patients with HGSOC.

关键点

计算基于部位间纹理的异质性指标是可行的;HGSOC部位捕获的纹理相似度的指标与总体生存相关;异质性指标也与HGSOC的不完全手术切除有关。



Calculating inter-site texture-based heterogeneity metrics was feasible • Metrics capturing texture similarities across HGSOC sites were associated with overall survival • Heterogeneity metrics were also associated with incomplete surgical resection of HGSOC.

关键词

卵巢癌,放射基因组学,影像组学,生存,纹理



Ovarian cancer; Radiogenomics; Radiomics; Survival; Texture


慧影医疗科技(北京)有限公司

地点:北京市海淀区中关村东升科技园B2-C103

电话:400-890-9020

邮箱:radcloud@huiyihuiying.com

关闭
图片
图片
  • 人工智能诊断云平台