Share：
Share this content in WeChat
X
[Chinese][PDF]107170
• Clinical Article •
A neural network radiomics model for diagnosing lymph node metastasis in cervical cancer
LING Rennan  YANG Ruofeng  YI Qinqin  RAO Zibin  YANG Yi  JIN Hongtao  CHENG Lixin 

Cite this article as: Ling RN, Yang RF, Yi QQ, et al. A neural network radiomics model for diagnosing lymph node metastasis in cervical cancer[J]. Chin J Magn Reson Imaging, 2021, 12(10): 16-21. DOI:10.12015/issn.1674-8034.2021.10.004.


[Abstract] Objective To construct a noninvasive multiparametric MRI-based radiomics model to predict lymph node metastasis in cervical cancer. Materials andMethods One hundred and seventy-eight cases of cervical cancer were analyzed retrospectively. While 9 clinical-pathological features were extracted, and 3 features were extracted into the model by analysis of variance (ANOVA). The volume of interest (VOI) was drawn by two observers respectively. Finally, 428 radiomics features of every tumor sample were extracted. Radiomics characteristics combined with various clinical-pathological characteristics were used to build the models of 428, 437 and 431 dimensions. The neural network model and support vector machine model are constructed and evaluated by torch and sklearn of Python library. Intraclass correlation coefficient (ICC) was used to evaluate inter observer reliability. Classification accuracy, sensitivity, specificity and area under the receiver operating characteristics curve (AUC) were used to measure the performance of the model. Youden index evaluated the performance of a dichotomous diagnostic test and the authenticity, by metrics.roc in sklearn_ROC curve.Results The average ICC was 0.819, which was highly reliable of the radiomics features performed by the two radiologists. The average ICC was 0.796, which was highly reproducibility of the radiomics features performed by one radiologist. The AUC of 431 dimension neural network model in test was 0.882, and the accuracy, sensitivity and specificity were 0.810, 0.840 and 0.741 respectively, which are better than other models.Conlusions The multiparametric MRI radiomics model based on neural network can be leveraged as a noninvasive diagnosis marker to predict LMN in cervical cancer.
[Keywords] cervical cancer；lymph node；metastasis；radiomics；machine learning

LING Rennan1   YANG Ruofeng2   YI Qinqin1   RAO Zibin1   YANG Yi3   JIN Hongtao4   CHENG Lixin5*  

1 Department of Radiology, Shenzhen People's Hospital, 2nd Clinical Medical College of Jinan University, 1st Affiliated Hospital of Southern University of Science and Technology, Shenzhen 518020, China

2 John Hopcroft Center for Computer Science, Shanghai Jiao Tong University, Shanghai 200240, China

3 Department of Gynecology,Shenzhen People's Hospital, 2nd Clinical Medical College of Jinan University, 1st Affiliated Hospital of Southern University of Science and Technology, Shenzhen 518020, China

4 Department of Pathology, Shenzhen People's Hospital, 2nd Clinical Medical College of Jinan University, 1st Affiliated Hospital of Southern University of Science and Technology, Shenzhen 518020, China

5 Department of Critical Care Medicine, Shenzhen People's Hospital, 2nd Clinical Medical College of Jinan University, 1st Affiliated Hospital of Southern University of Science and Technology, Shenzhen 518020, China

Cheng LX, E-mail: easonlcheng@gmail.com

Conflicts of interest   None.

ACKNOWLEDGMENTS Medical Scientific Research Foundation of Guangdong Province of China (NO. B2020004)
Received  2021-05-07
Accepted  2021-07-01
DOI: 10.12015/issn.1674-8034.2021.10.004
Cite this article as: Ling RN, Yang RF, Yi QQ, et al. A neural network radiomics model for diagnosing lymph node metastasis in cervical cancer[J]. Chin J Magn Reson Imaging, 2021, 12(10): 16-21. DOI:10.12015/issn.1674-8034.2021.10.004.

[1]
Koh WJ, Abu-Rustum NR, Bean S, et al. Cervical Cancer, Version 3.2019, NCCN Clinical Practice Guidelines in Oncology[J]. J Natl Compr Canc Netw: JNCCN, 2019, 17(1): 64-84. DOI: 10.6004/jnccn.2019.0001.
[2]
Siegel RL, Miller KD, Fuchs HE, et al. Cancer Statistics, 2021[J]. CA Cancer J Clin, 2021, 71(1): 7-33. DOI: 10.3322/caac.21654.
[3]
Bourgioti C, Chatoupis K, Moulopoulos LA. Current imaging strategies for the evaluation of uterine cervical cancer[J]. World J Radiol, 2016, 8(4): 342-354. DOI: 10.4329/wjr.v8.i4.342.
[4]
Rossi EC, Tanner E. Controversies in sentinel lymph node biopsy for gynecologic malignancies[J]. J Minim Invasive Gynecol, 2021, 28(3): 409-417. DOI: 10.1016/j.jmig.2020.12.025.
[5]
Cui YQ, Wang LL, Zhao LP, et al. Application progress of radiomics in cervical cancer[J]. Chin J Magn Reson Imaging, 2020, 11(6): 477-480. DOI: 10.12015/issn.1674-8034.2020.06.020.
[6]
Kan YY, Dong D, Zhang YC, et al. Radiomic signature as a predictive factor for lymph node metastasis in early-stage cervical cancer[J]. J Magn Reson Imaging, 2019, 49(1): 304-310. DOI: 10.1002/jmri.26209.
[7]
Li K, Sun H, Lu Z, et al. Value of [18F] FDG PET radiomic features and VEGF expression in predicting pelvic lymphatic metastasis and their potential relationship in early-stage cervical squamous cell carcinoma[J]. Eur J Radiol, 2018, 106: 106-166. DOI: 10.1016/j.ejrad.2018.07.024.
[8]
Wang YB, Liao ZH, Li H. A comparative study of diffusion-weighted imaging region of interest selection in cervicalcancer differentiation, lymph node metastasis and lymphovascular invasion[J]. Tianjin Medical Journal, 2021, 49(1): 79-84. DOI: 10.11958/20201290.
[9]
Wu QX, Wang S, Chen X, et al. Radiomics analysis of magnetic resonance imaging improves diagnostic performance of lymph node metastasis in patients with cervical cancer[J]. Radiother Oncol, 2019, 138: 141-148. DOI: 10.1016/j.radonc.2019.04.035.
[10]
Ferrandina G, Anchora LP, Gallotta V, et al. Can we define the risk of lymph node metastasis in early-stage cervical cancer patients? A large-scale, retrospective study[J]. Annals of Surgical Oncology, 2017, 24(8): 2311-2318. DOI: 10.1245/s10434-017-5917-0.
[11]
Geeitha S, Thangamani M. Incorporating EBO-HSIC with SVM for gene selection associated with cervical cancer classification[J]. J Med Syst, 2018, 42(11): 225. DOI: 10.1007/s10916-018-1092-5.
[12]
Luo Y, Mei DD, Gong JS, et al. Multiparametric MRI-based radiomics nomogram for predicting lymphovascular space invasion in endometrial carcinoma[J]. J Magn Reson Imaging, 2020, 52(4): 1257-1262. DOI: 10.1002/jmri.27142.
[13]
Chen XX, Liu W, Thai TC, et al. Developing a new radiomics-based CT image marker to detect lymph node metastasis among cervical cancer patients[J]. Comput Methods Programs Biomed, 2020, 197: 105759. DOI: 10.1016/j.cmpb.2020.105759.
[14]
Limkin EJ, Sun R, Dercle L, et al. Promises and challenges for the implementation of computational medical imaging (radiomics) in oncology[J]. Ann Oncol, 2017, 28(6): 1191-1206. DOI: 10.1093/annonc/mdx034.
[15]
Dong TT, Yang C, Cui BX, et al. Development and validation of a deep learning radiomics model predicting lymph node status in operable cervical cancer[J]. Front Oncol, 2020, 10: 464. DOI: 10.3389/fonc.2020.00464.
[16]
Brinker TJ, Hekler A, Enk AH, et al. Deep neural networks are superior to dermatologists in melanoma image classification[J]. Eur J Cancer, 2019, 119:11-17. DOI: 10.1016/j.ejca.2019.05.023.
[17]
Gulshan V, Peng L, Coram M, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs[J]. JAMA, 2016, 316(22): 2402-2410. DOI: 10.1001/jama.2016.17216.
[18]
Ju ZJ, Guo W, Gu SS, et al. CT based automatic clinical target volume delineation using a dense-fully connected convolution network for cervical Cancer radiation therapy[J]. BMC Cancer, 2021, 21(1): 243. DOI: 10.1186/s12885-020-07595-6.
[19]
Chen JG, Bai PG, Zhang XC, et al. Automated intensity modulated radiation therapy treatment planning for cervical cancer based on convolution neural network[J]. Technol Cancer Res Treat, 2020, 19: 1533033820957002. DOI: 10.1177/1533033820957002.
[20]
Tan XY, Li KX, Zhang JC, et al. Automatic model for cervical cancer screening based on convolutional neural network: a retrospective, multicohort, multicenter study[J]. Cancer Cell Int, 2021, 21(1): 35. DOI: 10.1186/s12935-020-01742-6.
[21]
Lu Y, Peng WW, Song J, et al. On the potential use of dynamic contrast-enhanced (DCE) MRI parameters as radiomic features of cervical cancer[J]. Med Phys, 2019, 46(11): 5098-5109. DOI: 10.1002/mp.13821.
[22]
Yu Z, Zhao ZH, Wei MZ, et al. Dynamic contrast-enhanced magnetic resonance imaging in pathological grading of cervical cancer: Extended Tofts Linear model versus Extended Tofts model[J]. Zhejiang Medical Journal, 2021, 43(2): 148-152. DOI: 10.12056/j.issn.1006-2785.2021.43.2.2020-1643.

PREV The value of T2WI sequence-based radiomics in predicting recurrence of acute pancreatitis
NEXT The value of enhanced T2 star weighted angiography in the diagnosis of cervical cancer and the evaluation of radiotherapy effect
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn