Bayesian network-based risk stratification of allergic sensitization using NHANES population data
4 affiliations
- 1Department of Artificial Intelligence and BigData, Al-Farabi Kazakh National University, Almaty, Kazakhstan
- 2Higher School of Information Technology and Engineering, Astana International University, Astana, Kazakhstan
- 3Department of IT and Service, «Q» University, Almaty, Kazakhstan
- 4Department of Information Systems, International University of Information Technologies, Almaty, Kazakhstan
Abstract
Allergic sensitization is commonly modeled as a binary classification outcome, an approach that discards the probabilistic structure linking demographic, immunological, and clinical variables and offers limited support for individualized risk communication. This work develops and evaluates a Bayesian network (BN) for probabilistic stratification of laboratory-defined serological IgE sensitization, fitted on harmonized data from the National Health and Nutrition Examination Survey (NHANES) 2005–2006 cycle (analytic cohort n = 8,285). The network integrates demographic covariates (age, sex), immunological markers (total serum IgE, eosinophil percentage), and clinical indicators (physician-diagnosed asthma, self-reported allergic symptoms) into a directed acyclic graph whose conditional probability tables were estimated by maximum-likelihood estimation with Laplace smoothing. Serological IgE sensitization, defined as at least one allergen-specific IgE ≥ 0.35 kUA/L across a 19-allergen panel, was present in 47.1% of the cohort. A five-fold cross-validated comparison of BN, logistic regression (LR), and random forest (RF) on identical predictor sets showed indistinguishable discrimination: AUC 0.808/0.809/0.808 (IgE + eosinophils) and 0.820/0.824/0.822 (IgE + eosinophils + age + sex). Individual-level calibration was excellent: Brier score = 0.171, calibration slope = 0.9998 (95% CI 0.9563–1.0434), intercept = −0.0001 (95% CI − 0.0522 to 0.0520). The three posterior probability strata (low < 0.40, intermediate 0.40–0.70, high > 0.70) correspond structurally to the three total IgE tertiles; adding eosinophil category improves continuous discrimination by ΔAUC = +0.010 but reclassifies zero participants between strata (categorical NRI = 0). Bidirectional inference—updating sensitization probability from a confirmed asthma diagnosis—raised the posterior from 0.472 to 0.651 (ΔAUC +0.006, asthma subsample n = 8,275), confirming a structural advantage unavailable from unidirectional classifiers. Observed serological sensitization rates across strata were 13.8% (low), 46.1% (intermediate), and 81.6% (high). The framework is offered as a transparent, auditable methodological demonstration of probabilistic modeling of serological IgE sensitization. It does not model clinical allergy and should not be used to guide clinical decisions without prospective validation. Its principal advantages over discriminative classifiers are structural: bidirectional inference, modular auditability of conditional probability tables, and native uncertainty propagation.
Related
- Could Dirofilaria immitis contribute to occult allergic sensitization? Serological evidence from allergic patients in a hyperendemic area
- Neonatal IL-4 and IL-5 signaling drives long-term inflammation after early-life allergen exposure in a mouse model of allergic sensitization
- Retrospectively reported indoor exposures, allergic multimorbidity, and questionnaire-based stratification among children aged 3–8 years in Northwest China: a cross-sectional study
- Patterns of Allergen Sensitization in Patients with Allergic Disease in Wenzhou, China: A Cross-Sectional Study
- From cartilage degeneration to pain sensitization: a multi-tissue mitochondrial dysfunction network in osteoarthritis
BibTeX
@article{shayakhmetova2026bayesian,
title = {{Bayesian network-based risk stratification of allergic sensitization using NHANES population data}},
author = {Assem Shayakhmetova and Nurbolat Tasbolatuly and Anar Sultangaziyeva and Marat Shurenov and Aigul Bissarinova and Nurbol Kurishbay},
journal = {Frontiers in Artificial Intelligence},
year = {2026},
month = oct,
doi = {10.3389/frai.2026.1920823},
url = {https://doi.org/10.3389/frai.2026.1920823},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Bayesian network-based risk stratification of allergic sensitization using NHANES population data
AU - Assem Shayakhmetova
AU - Nurbolat Tasbolatuly
AU - Anar Sultangaziyeva
AU - Marat Shurenov
AU - Aigul Bissarinova
AU - Nurbol Kurishbay
PY - 2026
DA - 2026/10/21
JO - Frontiers in Artificial Intelligence
DO - 10.3389/frai.2026.1920823
UR - https://doi.org/10.3389/frai.2026.1920823
AB - Allergic sensitization is commonly modeled as a binary classification outcome, an approach that discards the probabilistic structure linking demographic, immunological, and clinical variables and offers limited support for individualized risk communication. This work develops and evaluates a Bayesian network (BN) for probabilistic stratification of laboratory-defined serological IgE sensitization, fitted on harmonized data from the National Health and Nutrition Examination Survey (NHANES) 2005–2006 cycle (analytic cohort n = 8,285). The network integrates demographic covariates (age, sex), immunological markers (total serum IgE, eosinophil percentage), and clinical indicators (physician-diagnosed asthma, self-reported allergic symptoms) into a directed acyclic graph whose conditional probability tables were estimated by maximum-likelihood estimation with Laplace smoothing. Serological IgE sensitization, defined as at least one allergen-specific IgE ≥ 0.35 kUA/L across a 19-allergen panel, was present in 47.1% of the cohort. A five-fold cross-validated comparison of BN, logistic regression (LR), and random forest (RF) on identical predictor sets showed indistinguishable discrimination: AUC 0.808/0.809/0.808 (IgE + eosinophils) and 0.820/0.824/0.822 (IgE + eosinophils + age + sex). Individual-level calibration was excellent: Brier score = 0.171, calibration slope = 0.9998 (95% CI 0.9563–1.0434), intercept = −0.0001 (95% CI − 0.0522 to 0.0520). The three posterior probability strata (low < 0.40, intermediate 0.40–0.70, high > 0.70) correspond structurally to the three total IgE tertiles; adding eosinophil category improves continuous discrimination by ΔAUC = +0.010 but reclassifies zero participants between strata (categorical NRI = 0). Bidirectional inference—updating sensitization probability from a confirmed asthma diagnosis—raised the posterior from 0.472 to 0.651 (ΔAUC +0.006, asthma subsample n = 8,275), confirming a structural advantage unavailable from unidirectional classifiers. Observed serological sensitization rates across strata were 13.8% (low), 46.1% (intermediate), and 81.6% (high). The framework is offered as a transparent, auditable methodological demonstration of probabilistic modeling of serological IgE sensitization. It does not model clinical allergy and should not be used to guide clinical decisions without prospective validation. Its principal advantages over discriminative classifiers are structural: bidirectional inference, modular auditability of conditional probability tables, and native uncertainty propagation.
SN - 2624-8212
ER -