Skip to content
DraconiticResearch
Search Sources API
182,544 studies31 sourcesUpdated 6 min ago
Peer-reviewedCC BYScientific Reports

Machine learning-based analysis of risk factors and construction of a predictive model for hyperuricemia in Chinese health examination population

  • ChuXia Tan1
  • Xue He1
  • Lijun Li2
  • Ying Li1
  • Pingting Yang1
  • Yuxuan Li3
  • Juan Luo3
  • Mingyue Huang3
  • Honglian Zhang4
  • Andy S. K. Cheng5
5 affiliations
  1. 1Health Management Medicine Center, The Third Xiangya Hospital, Central South University
  2. 2School of Psychological Sciences, Monash University
  3. 3Xiangya Nursing School of Central South University
  4. 4Nursing Department, The Third Xiangya Hospital, Central South University
  5. 5School of Health Sciences, Western Sydney University
Read full textPDFDOICite

Abstract

Hyperuricemia (HUA) imposes a growing public health burden, calling for better risk stratification tools. In this cross-sectional study of 4906 Chinese adults undergoing routine health checks (overall HUA prevalence: 26.0%), we built machine learning-based predictive models using a stratified 80/20 data split. To avoid variable selection bias, we applied LASSO regression with tenfold cross-validation, which identified 12 core predictors from routine clinical and demographic data. Among four algorithms tested, the Gradient Boosting (GB) model showed the best discrimination (AUC = 0.770) and good calibration (slope = 0.982, intercept = − 0.007, Brier score = 0.157). SHAP analysis revealed serum creatinine (Scr), HDL cholesterol (HDL-C), and body mass index (BMI) as the top predictors. Notably, SHAP interaction plots uncovered a nonlinear rise in risk above a Scr threshold and a compounded risk when high Scr coincided with low HDL-C. In summary, the well-calibrated GB model offers a reliable, data-driven tool for HUA risk screening, with insights into marker interactions to guide targeted prevention and early intervention.

Related

  1. Helicobacter pylori infection, hyperuricemia, and serum uric acid levels: The role of metabolic, renal, and inflammatory factors in a large health examination population
    Zhiyue Xu, Xiuyu Wang, Jie Huang +2
    PLOS OnePeer-reviewedCC BY 4.0Full textPhysiologyPLOS
  2. Toward a Predictive Model for Medical Residency Entrance Examination Performance in Brazil: A Preregistered Cross-Sectional Study and Exploratory Prototype Score Calculator
    Lucas R Nunes, Mariana C Rocha
    CureusPeer-reviewedCC BYFull textGeneral medicineEurope PMC
  3. Construction and validation of a predictive nomogram for vascular crisis risk following free flap surgery
    Yan-liang Qu, Juan Xue, Xing-yu Du +4
    Frontiers in medicinePeer-reviewedCC BYFull textGeneral medicineEurope PMC
  4. Association between smoking status and choroidal thickness in the Korean population: Korea National Health and Nutrition Examination Survey 2017–2021
    Jae Shin Song, Min Seok Kim
    PLOS OnePeer-reviewedCC BY 4.0Full textEndocrinology & metabolismPhysiologyPLOS
  5. TyG-CVAI for identifying hyperuricemia: geographical external validation, model updating, and explainable machine learning
    Chen Liang, Zhengyang Lu, Zhengtao Li +1
    Frontiers in EndocrinologyPeer-reviewedCC BY 4.0Endocrinology & metabolismCrossref

Cite

BibTeXRIS
BibTeX
@article{tan2026machine,
  title = {{Machine learning-based analysis of risk factors and construction of a predictive model for hyperuricemia in Chinese health examination population}},
  author = {ChuXia Tan and Xue He and Lijun Li and Ying Li and Pingting Yang and Yuxuan Li and Juan Luo and Mingyue Huang and Honglian Zhang and Andy S. K. Cheng},
  journal = {Scientific Reports},
  year = {2026},
  month = jul,
  doi = {10.1038/s41598-026-62552-w},
  url = {https://doi.org/10.1038/s41598-026-62552-w},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Machine learning-based analysis of risk factors and construction of a predictive model for hyperuricemia in Chinese health examination population
AU  - ChuXia Tan
AU  - Xue He
AU  - Lijun Li
AU  - Ying Li
AU  - Pingting Yang
AU  - Yuxuan Li
AU  - Juan Luo
AU  - Mingyue Huang
AU  - Honglian Zhang
AU  - Andy S. K. Cheng
PY  - 2026
DA  - 2026/07/01
JO  - Scientific Reports
DO  - 10.1038/s41598-026-62552-w
UR  - https://doi.org/10.1038/s41598-026-62552-w
AB  - Hyperuricemia (HUA) imposes a growing public health burden, calling for better risk stratification tools. In this cross-sectional study of 4906 Chinese adults undergoing routine health checks (overall HUA prevalence: 26.0%), we built machine learning-based predictive models using a stratified 80/20 data split. To avoid variable selection bias, we applied LASSO regression with tenfold cross-validation, which identified 12 core predictors from routine clinical and demographic data. Among four algorithms tested, the Gradient Boosting (GB) model showed the best discrimination (AUC = 0.770) and good calibration (slope = 0.982, intercept = − 0.007, Brier score = 0.157). SHAP analysis revealed serum creatinine (Scr), HDL cholesterol (HDL-C), and body mass index (BMI) as the top predictors. Notably, SHAP interaction plots uncovered a nonlinear rise in risk above a Scr threshold and a compounded risk when high Scr coincided with low HDL-C. In summary, the well-calibrated GB model offers a reliable, data-driven tool for HUA risk screening, with insights into marker interactions to guide targeted prevention and early intervention.
SN  - 2045-2322
ER  -