Related Experiment Video
Updated: Oct 3, 2025
Probing the Limits of Egg Recognition Using Egg Rejection Experiments Along Phenotypic Gradients
Published on: August 22, 2018
Establishing a predictive model for the evaluation of fecundity
Qiong Zhan1,2, Jing Zhao3, Yasheng Paziliya2
1School of Public Health, Xinjiang Medical University, Urumqi, P. R. China.
Aim:
We aim to establish a predictive model for the evaluation of fecundity based on infertility-related factors.
Methods:
A total of 410 expectant couples who visited the First Affiliated Hospital of Xinjiang Medical University on January 1, 2017 and June 10, 2019 were included in this study. The 1-year follow-up was carried out to investigate the pregnancy of the female. They were divided into model group and test group, respectively. The basic information, life behavior and clinical indices were screened using the Logistics regression analysis and LASSO regression analysis. In addition, the multivariate logistic regression was used to establish the model for the prediction of fecundity risk.
Results:
The risk factors for the predictive model included female age and occupational pressure, gynecological disease, anti-Müllerian hormone (AMH), follicle-stimulating hormone (FSH), fasting plasma glucose (FPG), depression, as well as male smoking. The area under the curve (AUC) for the model A and model B was 0.954 (0.931 ~ 0.978) and 0.955 (0.931 ~ 0.979), respectively. The AUC in the test group was 0.917 (0.869 ~ 0.965) and 0.921 (0.873 ~ 0.968). There were no statistical differences in the fitting value and measured values in the model group.
Conclusions:
We established a predictive model for the evaluation of fecundity, which showed a satisfactory accuracy and discriminatory power.
