Optimizing NIPT Timing through Integrated Modeling: A GAMM and Risk-Based Approach for Personalized Prenatal Screening

View at publisher Summarise with VIRAT
Authors: Leyi Wang, Yeqi Zhang, Yurui Li, Shilin Gao
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study addresses the challenge of optimizing Non-Invasive Prenatal Testing (NIPT) timing by developing an integrated modeling framework that analyzes fetal Y-chromosome concentration dynamics. Current NIPT guidelines recommend fixed gestational windows, failing to account for individual maternal characteristics like BMI that significantly impact fetal DNA fraction. To overcome this limitation, we propose a multi-phase methodology combining Generalized Additive Mixed Models (GAMM) for relationship modeling and enhanced risk quantification with parallel Monte Carlo simulation for optimization. The experimental process first compares GAMM with Random Forest and LASSO regression to establish the optimal relationship model, then constructs a comprehensive risk function incorporating time pressure, multi-factor deviation, grouping uncertainty, and concentration-related risks. The results demonstrate that GAMM achieves superior performance with R2=0.401, significantly outperforming comparative methods. The optimization yields personalized testing schedules for five BMI subgroups, with compliance rates ranging from 64.0% to 71.2%. This research provides a data-driven solution for NIPT timing optimization, demonstrating the advantages of integrated risk modeling in handling complex maternal-fetal parameters for personalized prenatal screening.

Fields of Study & Subjects

Prenatal Screening and Diagnostics Maternal and Perinatal Health Interventions Artificial Intelligence in Healthcare

Publication details

Journal / source
IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
Pages
1191–1196
Publication date
28 January 2026
Language
EN