Radiomics analysis on T2-MR image to predict lymphovascular space invasion in cervical cancer

Shou Wang; Xi Chen; Zhenyu Liu; Qingxia Wu; Yongbei Zhu; Meiyun Wang; Jie Tian

doi:10.1117/12.2513129

13 March 2019 Radiomics analysis on T2-MR image to predict lymphovascular space invasion in cervical cancer

Shou Wang, Xi Chen, Zhenyu Liu, Qingxia Wu, Yongbei Zhu, Meiyun Wang, Jie Tian

Proceedings Volume 10950, Medical Imaging 2019: Computer-Aided Diagnosis; 1095040 (2019) https://doi.org/10.1117/12.2513129
Event: SPIE Medical Imaging, 2019, San Diego, California, United States

Abstract

Lymphovascular space invasion (LVSI) is an important determinant for selecting treatment plan in cervical cancer (CC). For CC patients without LVSI, conization is recommended; otherwise, if LVSI is observed, hysterectomy and pelvic lymph node dissection are required. Despite the importance, current identification of LVSI can only be obtained by pathological examination through invasive biopsy or after surgery. In this study, we provided a non-invasive and preoperative method to identify LVSI by radiomics analysis on T2-magnetic resonance image (MRI), aiming at assisting personalized treatment planning. We enrolled 120 CC patients with T2 image and clinical information, and allocated them into a training set (n = 80) and a testing set (n= 40) according to the diagnostic time. Afterwards, 839 image features were extracted to reflect the intensity, shape, and high-dimensional texture information of CC. Among the 839 radiomic features, 3 features were identified to be discriminative by Least absolute shrinkage and selection operator (Lasso)-Logistic regression. Finally, we built a support vector machine (SVM) to predict LVSI status by the 3 radiomic features. In the independent testing set, the radiomics model achieved area under the receiver operating characteristic curve (AUC) of 0.7356, classification accuracy of 0.7287. The radiomics signature showed significant difference between non-LVSI and LVSI patients (p<0.05). Furthermore, we compared the radiomics model with clinical model that uses clinical information, and the radiomics model showed significant improvement than clinical factors (AUC=0.5967 in the validation cohort for clinical model).

Citation Download Citation

Shou Wang, Xi Chen, Zhenyu Liu, Qingxia Wu, Yongbei Zhu, Meiyun Wang, and Jie Tian "Radiomics analysis on T2-MR image to predict lymphovascular space invasion in cervical cancer", Proc. SPIE 10950, Medical Imaging 2019: Computer-Aided Diagnosis, 1095040 (13 March 2019); https://doi.org/10.1117/12.2513129

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Cervical cancer

Tumors

Image analysis

Feature extraction

Magnetic resonance imaging

Feature selection

Medical imaging

Show All Keywords

Keywords/Phrases

Search In:

Publication Years