于媛媛，中共党员，流行病与卫生统计学博士，副研究员。主要研究方向：观察性研究中因果推断的理论方法及应用研究，在大数据背景下，解决复杂数据中的未知混杂控制、提高疾病预测的精确性与稳健性，推动精准医疗和个体化治疗的发展，为公共卫生和临床干预提供了科学依据和技术支持。主要研究内容包括多中心重复测量数据未知混杂控制的因果推断方法研究；稳健疾病预测预警模型研究；基于统计遗传学的方法学研究和临床治疗及干预研究等。
1. 大数据真实世界研究
2. 因果推断
1. Zhang S, … Yu Y#, Xue F#.. Interpretable machine learning model for digital lung cancer prescreening in Chinese populations with missing data.. NPJ Digital Medicine, 2024.
2. Hou L+, Yu Y+, et al. Demystifying inconsistent two-sample mendelian randomization estimations using selection diagram. BMC Medical Research Methodology, 2025.
3. Xu, H.+, Yu, Y.+,. Precision lung cancer screening from CT scans using a VGG16-based convolutional neural network.. Frontiers in oncology, 2024.
4. Cui Z+, Ma Y+, Yu Y+, et al.. Short-term exposure to ambient fine particulate pollution aggravates ventilator-associated pneumonia in pediatric intensive care patients undergoing cardiovascular surgeries.. Environmental health., 2023.
5. Yu Y+, Hou L+,, et al.. Impact of nonrandom selection mechanisms on the causal effect estimation for two-sample Mendelian randomization methods.. PLoS Genetics, 2022.
6. Yu Y, Li H,, et al.. Identification and estimation of causal effects using a negativecontrol exposure in time-series studies with applications to environmental epidemiology.. American Journal of Epidemiology, 2021.
7. Yu Y, Li H, Sun X, et al.. The alarming problems of confounding equivalence using logistic regression models in the perspective of causal diagrams.. BMC Medical Research Methodology, 2017.
1. 基于纵向数据的未知混杂控制及双向因果关系推断方法研究
2. 多中心重复测量数据未知混杂控制的因果推断方法研究, 国家自然科学基金-2027-12-31
