BMI-Stratified Mixed-Effects Modeling of Fetal DNA Fraction for NIPT Timing Optimization and Ensemble-Based Female Fetal

Abstract

Non-invasive prenatal testing (NIPT) has gained widespread clinical adoption as an efficient and precise method for screening fetal chromosomal abnormalities. However, there remains room for optimization in balancing the timing of testing with diagnostic accuracy. This study employs multilevel modeling of clinical data from 267 pregnant women to propose a personalized testing timing optimization strategy based on multiple factors, including fetal cell-free DNA concentration and maternal body mass index (BMI). Using a mixed-effects model, we analyzed the relationship between BMI, gestational age, GC content, and fetal Y chromosome concentration, thereby establishing a predictive framework for optimal testing timing across different BMI groups. Experimental results indicate optimal testing timepoints of 10.0 weeks for low-BMI groups, 13.0 weeks for moderate-BMI groups, and 15.8 weeks for high-BMI groups. Compared to conventional methods, the model enables testing 3 to 6 weeks earlier. For determining chromosomal abnormalities in female fetuses, integrating multivariate Gaussian discriminant analysis with the RUSBoost ensemble learning algorithm successfully enhanced detection sensitivity and accuracy. Particularly when combining Z-scores with GC content, the approach effectively distinguished between normal and abnormal fetuses. Comprehensive analysis indicates this study provides scientific evidence for personalized optimization of NIPT testing timing and accurate diagnosis of chromosomal abnormalities in female fetuses.