Delta-radiomics特征用于预测非小细胞肺癌患者的结局(英文)

2017-04-03 15:38:48 admin 3

Delta-radiomics features for the prediction of patient outcomes in non-small cell lung cancer.

Sci Rep. 2017 Apr 3

Fave X1,2, Zhang L3, Yang J3, Mackin D3, Balter P3, Gomez D4, Followill D3, Jones AK5, Stingo F6, Liao Z4, Mohan R3, Court L3,7.

Author information

    1.Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX, 77030, USA. xjfave@mdanderson.org.

    2.The University of Texas Graduate School of Biomedical Sciences at Houston, 6767 Bertner Ave, Houston, TX, 77030, USA. xjfave@mdanderson.org.

    3.Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX, 77030, USA.

    4.Department of Radiation Oncology, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX, 77030, USA.

    5.Department of Imaging Physics, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX, 77030, USA.

    6.Dipartimento Di Statistica, Informatica, Applicazioni, University of Florence, Viale Morgagni, 59, Florence, 50134, Italy.

    7.The University of Texas Graduate School of Biomedical Sciences at Houston, 6767 Bertner Ave, Houston, TX, 77030, USA.

Abstract

Radiomics is the use of quantitative imaging features extracted from medical images to characterize tumor pathology or heterogeneity. Features measured at pretreatment have successfully predicted patient outcomes in numerous cancer sites. This project was designed to determine whether radiomics features measured from non-small cell lung cancer (NSCLC) change during therapy and whether those features(delta-radiomics features) can improve prognostic models. Features were calculated from pretreatment and weekly intra-treatment computed tomography images for 107 patients with stage III NSCLC. Pretreatment images were used to determine feature-specific image preprocessing. Linear mixed-effects models were used to identify features that changed significantly with dose-fraction. Multivariate models were built for overall survival, distant metastases, and local recurrence using only clinical factors, clinical factors and pretreatment radiomics features, and clinical factors, pretreatment radiomics features, and delta-radiomics features. All of the radiomics features changed significantly during radiation therapy. For overall survival and distant metastases, pretreatment compactness improved the c-index. For local recurrence, pretreatment imaging features were not prognostic, while texture-strength measured at the end of treatment significantly stratified high- and low-risk patients. These results suggest radiomics features change due to radiation therapy and their values at the end of treatment may be indicators of tumor response.


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