Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Advances in Computer, Signals and Systems
  4. Vol 10, Issue 1, 2026

Optimization of NIPT Timing Using Random Forest and Hierarchical Clustering

Download as PDF

DOI: 10.23977/acss.2026.100118 | Downloads: 6 | Views: 73

Author(s)

Weiyi Zhu 1, Can Wu 1, Hongxun Ye 1, Shengjun Xu 1

Affiliation(s)

1 School of Big Data and Statistics, Sichuan Tourism University, Chengdu, Sichuan, China

Corresponding Author

Weiyi Zhu

ABSTRACT

In this study, a three-step progressive algorithm framework is constructed to realize the individualized optimization of NIPT time points. In the first step, based on polynomial regression and gradient lifting tree, the nonlinear relationship between Y chromosome concentration, gestational age and BMI was revealed, and the gradient lifting tree was optimally fitted (test set R²=0.2521). In the second step, K-means clustering and regression trees were used to adaptively group male BMI, and the optimal NIPT time points of each subgroup were determined by Monte Carlo simulation error analysis, and the optimal grouping intervals were [26.62, 29.90], [29.96, 32.19], [32.26, 34.93], and [35.06, 39.14], and the corresponding time points were 16, 19, 17, and 24 weeks, respectively, and the confidence interval width was less than 1.2 weeks. In the third step, multiple factors such as height, weight, and age were introduced, and a random forest regression model was constructed to predict the gestational age of the first standard (test R²=0.942) by introducing multiple factors such as height, weight, and age, and the optimal time points of the four groups were 16.8, 18.8, 18.9, and 24.0 weeks, respectively. The algorithm chain integrates nonlinear regression, cluster integration and random forest, which provides a data-driven quantitative basis for NIPT point-in-time decision-making.

KEYWORDS

Gradient Boosting Trees; K-means Clustering; Random Forest; Hierarchical Clustering; Monte Carlo Simulation; NIPT Timing Optimization

CITE THIS PAPER

Weiyi Zhu, Can Wu, Hongxun Ye, Shengjun Xu. Optimization of NIPT Timing Using Random Forest and Hierarchical Clustering. Advances in Computer, Signals and Systems (2026). Vol. 10, No. 1, 151-158. DOI: http://dx.doi.org/10.23977/acss.2026.100118.

REFERENCES

[1] Shi Y, He J, Liu H, Wang Y, Chen Z. Machine learning for predicting fetal DNA fraction in noninvasive prenatal screening[J]. BMC Bioinformatics, 2021, 22(1): 512.
[2] Wang L, Zhang X, Chen Y, Liu M. Gradient boosting for optimizing screening timing in noninvasive prenatal testing[J]. Frontiers in Genetics, 2022, 13: 894567.
[3] Chen X, Li J, Wang Y. Cluster analysis of maternal BMI and its association with fetal aneuploidy risk[J]. Journal of Perinatal Medicine, 2023, 51(4): 512-520.
[4] Zhang H, Liu Q, Zhou J. Hierarchical clustering for personalized NIPT timing recommendation[J]. IEEE Journal of Biomedical and Health Informatics, 2024, 28(2): 1123-1132.
[5] Liu J, Wang Y, Zhao S. Monte Carlo simulation-based robustness evaluation for prenatal screening models[J]. Computer Methods and Programs in Biomedicine, 2023, 231: 107401.
[6] Kim S Y, Park J H, Lee K. Deep learning-based noninvasive prenatal testing for fetal chromosome abnormalities[J]. Nature Communications, 2022, 13(1): 4562.
[7] Zhou Q, Yang L, Zhang R. Interpretable machine learning for gestational age prediction using multi-omics data[J]. Briefings in Bioinformatics, 2023, 24(3): bbad102.
[8] Huang T, Wang H, Chen S. Ensemble learning for detecting fetal subchromosomal abnormalities[J]. Human Genetics, 2022, 141(5): 1013-1024.
[9] Patel A, Gupta R, Sharma N. The role of maternal BMI in cell-free fetal DNA concentration: a systematic review[J]. Prenatal Diagnosis, 2021, 41(8): 952-961.
[10] Yang Z, Liu Y, Chen W. Feature selection and classification for NIPT data using SVM and random forest[J]. BMC Medical Genomics, 2024, 17(1): 34.
[11] Li M, Wang J, Zhang Y. Explainable artificial intelligence for prenatal diagnosis: a survey[J]. Artificial Intelligence in Medicine, 2024, 148: 102777.
[12] Gao Y, Xu L, Chen H. Uncertainty-aware deep learning for NIPT timing optimization[J]. IEEE Transactions on Biomedical Engineering, 2024, 71(5): 1602-1612.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 46386
Visits: 1053813

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.