Research on Dynamic Timing Decision-Making for Non-Invasive Prenatal Testing Based on Risk-Error Co-Optimization

Read full text Summarise with VIRAT
Authors: Yanting Liu
Cited by: 0
References: 10

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

Addressing the risks faced by high-BMI pregnant women in non-invasive prenatal testing (NIPT)—namely delayed attainment of sufficient fetal cell-free DNA (cfDNA) concentration and missed optimal intervention windows—this study establishes a data-driven system for optimizing testing timepoints. The research first preprocesses raw samples by converting gestational age into continuous numerical variables and develops evaluation metrics encompassing clinical intervention risks and testing accuracy. To address the limitation of static grouping in distinguishing individual differences, the K-means clustering algorithm was employed to reclassify pregnant women into five mutually exclusive clusters based on BMI characteristics. Kaplan-Meier survival analysis was then applied to reveal the temporal evolution of Y chromosome concentration attainment rates across different BMI levels. Building upon this foundation, a nonlinear optimization model was constructed to balance sequencing failure risk and missed therapeutic window risk. By minimizing total risk and sequencing error, optimal detection timepoints were determined for each group. Validation results demonstrated that the optimal detection timepoint for the high-BMI group was significantly delayed compared to the low-BMI group, with a maximum deviation of up to 4 weeks. Finally, sensitivity analysis via Monte Carlo simulation confirmed the model's robustness to sequencing perturbations

Fields of Study & Subjects

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

Publication details

Journal / source
International Journal of Public Health and Medical Research
Volume
6
Issue
2
Pages
51–61
Publication date
28 February 2026
Language
EN