Development of a Machine Learning Model for Risk Prediction of Ovarian Cancer
Principal Investigator
Name
Yixiang Zhang
Degrees
Ph.D
Institution
The Second Affiliated Hospital of Fujian Medical University
Position Title
Dr.
Email
About this CDAS Project
Study
PLCO
(Learn more about this study)
Project ID
PLCO-1192
Initial CDAS Request Approval
Apr 10, 2023
Title
Development of a Machine Learning Model for Risk Prediction of Ovarian Cancer
Summary
Ovarian cancer is a malignant tumor and ranks as the eighth most common cancer in women worldwide[1]. However, the challenge in the diagnosis of ovarian cancer arises due to its atypical early disease symptoms, and the non-availability of feasible screening strategies for this disease[1]. Although genetic-based ovarian cancer prediction models have been developed[2], their complexity and high cost make them unsuitable for widespread implementation. Serum CA-125 testing and vaginal ultrasound are important for its diagnosis and are suitable for most populations[3]. The present study aims to develop a simple and reliable risk prediction model for ovarian cancer through the use of CA-125, vaginal ultrasound, and other data obtained from a baseline questionnaire. The goal of this approach is to enhance early diagnosis of ovarian cancer and reduce the burden of disease.
[1]Varghese A, Lele S. Rare Ovarian Tumors. In: Lele S, ed. Ovarian Cancer. Brisbane (AU): Exon Publications; September 8, 2022.
[2]Dareng EO, Tyrer JP, Barnes DR, et al. Polygenic risk modeling for prediction of epithelial ovarian cancer risk [published correction appears in Eur J Hum Genet. 2022 Mar 22;:]. Eur J Hum Genet. 2022;30(3):349-362. doi:10.1038/s41431-021-00987-7
[3] Orr B, Edwards RP. Diagnosis and Treatment of Ovarian Cancer. Hematol Oncol Clin North Am. 2018;32(6):943-964. doi:10.1016/j.hoc.2018.07.010
[1]Varghese A, Lele S. Rare Ovarian Tumors. In: Lele S, ed. Ovarian Cancer. Brisbane (AU): Exon Publications; September 8, 2022.
[2]Dareng EO, Tyrer JP, Barnes DR, et al. Polygenic risk modeling for prediction of epithelial ovarian cancer risk [published correction appears in Eur J Hum Genet. 2022 Mar 22;:]. Eur J Hum Genet. 2022;30(3):349-362. doi:10.1038/s41431-021-00987-7
[3] Orr B, Edwards RP. Diagnosis and Treatment of Ovarian Cancer. Hematol Oncol Clin North Am. 2018;32(6):943-964. doi:10.1016/j.hoc.2018.07.010
Aims
1. To explore the correlation of CA-125, vaginal ultrasound, and other baseline questionnaire data with ovarian cancer risk.
2. To develop ovarian cancer risk prediction models using multiple machine learning algorithms.
3. To offer a straightforward and practical clinical tool for early diagnosis of ovarian cancer, allowing for disease prevention recommendations for patients.
Collaborators
Liuxia You, Chunmei Fan, Tebin Chen(Department of Laboratorial Medicine, The Second Affiliated Hospital of Fujian Medical University, Quanzhou)
Yifu Zeng (Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou)
Qingquan Chen, Kang Yang, Jiajing Zhuang,Ling Yao(Fujian Medical University, Fuzhou)