K-means Clustering Based on Self-Sampling and Multi-Risk Minimization in NIPT Screening

Liang Hu; Yeqi Zhang; Yanyan Wu; Qian Li

doi:10.23977/jaip.2026.090107

K-means Clustering Based on Self-Sampling and Multi-Risk Minimization in NIPT Screening

Download as PDF

DOI: 10.23977/jaip.2026.090107 | Downloads: 1 | Views: 61

Author(s)

Liang Hu ¹, Yeqi Zhang ¹, Yanyan Wu ^2,3, Qian Li ^2,4

Affiliation(s)

¹ Big Data College, Fuzhou University of International Studies and Trade, Fuzhou, Fujian, China
² College of Digital Technology and Engineering, Ningbo University of Finance and Economics, Ningbo, China
³ Faculty of Data Science, City University of Macau, Macau, China; Key Laboratory of Data Science and Intelligent Computing, Fuzhou University of International Studies and Trade, Fuzhou, Fujian, China
⁴ CKC Software Laboratory, Ningbo University, Ningbo, China

Corresponding Author

Yanyan Wu

ABSTRACT

This study aimed to optimize NIPT testing strategies and mitigate clinical risks associated with individual variability and detection errors. By focusing on pregnant women carrying male fetuses, it combined Bootstrap and K-means algorithms to develop the Bootstrap K-Means method, addressing grouping biases inherent in direct clustering. The study divided maternal BMI values into four optimal groups, establishing a total risk function, time-point expectations, failure rates, and sequencing quality risk weights. By solving for the optimal NIPT testing time points for each BMI group under constrained conditions, results indicated approximately 12.3 weeks for the first three groups, with the high-BMI group delayed to 13 weeks. Sensitivity analysis demonstrated the robustness of the model outcomes. An accuracy risk function incorporating cautionary factors (advanced maternal age, obesity, non-natural conception) was developed. The failure risk function model was modified and extended to analyze optimal testing timepoints for these factors. Results indicate higher accuracy risk and model sensitivity in the high-BMI group due to its greater obesity prevalence. This optimized mathematical model for NIPT testing provides quantitative evidence to support personalized, evidence-based prenatal screening strategies in clinical practice.

KEYWORDS

NIPT; BMI Grouping; K-means, Bootstrap; Risk Minimization

CITE THIS PAPER

Liang Hu, Yeqi Zhang, Yanyan Wu, Qian. K-means Clustering Based on Self-Sampling and Multi-Risk Minimization in NIPT Screening. Journal of Artificial Intelligence Practice (2026). Vol. 9, No. 1, 47-56. DOI: http://dx.doi.org/10.23977/jaip.2026.090107.

REFERENCES

[1] Xue Z ,Zhou A, Zhu X, et al. NIPT-PG: empowering non-invasive prenatal testing to learn from population genomics through an incremental pan-genomic approach [J]. Briefings in bioinformatics,2024,25(4).
[2] Fukuchi I J. Approximate Selection of All Populations Better Than a Control by Bootstrap[J]. Sankhya A, 2025,87(2):1-20.
[3] Wywiał J, Szkutnik T. Bootstrap tests for unbiasedness of predictors[J]. Communications in Statistics - Simulation and Computation,2025,54(6):1987-2004.
[4] FeinbergM, ClémençonS, RudazS, et al. Kernel‐Based Bootstrap Synthetic Data to Estimate Measurement Uncertainty in Analytical Sciences[J]. Journal of Chemometrics,2024,38(12):e3628.
[5] Wei Z, Tima M, Julie S, et al.Cost-effectiveness of prenatal screening and diagnostic strategies for Down syndrome: A microsimulation modeling analysis [J]. PloS one,2019,14(12):e0225281.
[6] Cogno N, Bauer R, Durante M . Mechanistic model of radiotherapy-induced lung fibrosis using coupled 3D agent-based and Monte Carlo simulations [J].Communications medicine,2024,4(1):16.
[7] Kwan Y L, Joo O K, Mi-Kyung L, et al. Estimating the measurement uncertainties of the international sensitivity index of 12 thromboplastins through Monte Carlo simulation [J]. Thrombosis Research,2023,224:32-37.
[8] Chen Jue, Zhu Liping, Cen Shuyuan, et al. Noninvasive prenatal screening using fetal cell-free DNA in maternal peripheral blood: A study of 21,092 cases [J]. Chinese Journal of Prenatal Diagnosis (Electronic Edition), 2023, 15(01): 29-34.
[9] You Zuping, Wang Zheng. Analysis of the Establishment of Institutions and Current Status of Prenatal Screening Using Fetal Cell-Free DNA in Peripheral Blood of Pregnant Women in Jinhua City [J]. Chinese Journal of Health Supervision, 2021, 28(02): 116-121.
[10] Zhai Chunya. Clinical Application Research of Non-Invasive Prenatal Testing (NIPT) in Sex Chromosome Abnormalities [D]. Tianjin Medical University,2021.
[11] Pang Yu. Clinical Application of Non-Invasive Prenatal Testing in Fetal Chromosomal Disease Screening [D]. Anhui Medical University, 2022.
[12] Tarasi L, Covelli M, Fatis D T C, et al. Prior Information Shapes Perceptual Confidence.[J].Journal of cognition,2025,8(1):11.

Subscription

E-Mail Alert

Downloads:	25532
Visits:	700046

K-means Clustering Based on Self-Sampling and Multi-Risk Minimization in NIPT Screening

Author(s)

Affiliation(s)

Corresponding Author

ABSTRACT

KEYWORDS

CITE THIS PAPER

REFERENCES

RESOURCES

JOIN US

PUBLICATION SERVICES

CONTACT US