Artificial Intelligence‐Based OSAS Biomarker Screening and Prediction Algorithm
IET Biometrics · 2026-01 · 2026 (1)
DOI: 10.1049/bme2/1251077 ↗Kurz gefasst
Fachartikel in der Zeitschrift IET Biometrics, online erschienen am 29.07.2026. Verfasst von Ting Lu und 1 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Traditional biomarker research has the problems of low manual screening efficiency, strong subjectivity, small sample size but high feature dimension, and difficulty in capturing complex relationships between variables, which limits its practical application in clinical practice. To solve the above problems, this article studies an obstructive sleep apnea syndrome (OSAS) biomarker screening and prediction algorithm based on artificial intelligence. LASSO (least absolute shrinkage and selection operator) regression and random forest feature importance ranking are used to automatically screen key biomarkers from high‐dimensional features, effectively improving screening efficiency and reducing subjective intervention. Combining regularized modeling strategies with cross‐validation techniques, the risk of overfitting can be suppressed, and the robustness of the model under small sample conditions can be improved. Gradient boosting trees and multilayer perceptron (MLP) models are also used to capture the nonlinear and interactive effects between variables, thereby building a more expressive prediction model. The experimental results show that the constructed model performs well in terms of accuracy, sensitivity, and computing performance, with a prediction accuracy (optimal parameter combination) of 0.86 and an inference time between 0.046 and 0.085 s, verifying the potential of artificial intelligence in OSAS biomarker mining and individualized risk prediction. This study provides adequate technical support for early screening and precision medicine of OSAS.
Zitieren
APA
Lu, T., & Wu, L. (2026). Artificial Intelligence‐Based OSAS Biomarker Screening and Prediction Algorithm. IET Biometrics, 2026(1). https://doi.org/10.1049/bme2/1251077
Vancouver
Lu T, Wu L. Artificial Intelligence‐Based OSAS Biomarker Screening and Prediction Algorithm. IET Biometrics. 2026;2026(1). doi:10.1049/bme2/1251077
BibTeX
@article{Lu2026,
author = {Lu, Ting and Wu, Liang},
title = {Artificial Intelligence‐Based OSAS Biomarker Screening and Prediction Algorithm},
journal = {IET Biometrics},
year = {2026},
volume = {2026},
number = {1},
doi = {10.1049/bme2/1251077}
}
EUVIMED