EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Calibrated Multi-Task Ensemble Architecture for Biomedical Risk Prediction

Zhainagul Khamitova, Gulmira Omarova, Madi Akhmetzhanov, Roza Burganova, Maksym Orynbassar, Umida Sabirova, Almagul Bukatayeva, Aliya Barakova, Gulnoz Jiyanmuratova, Dilchekhra Yuldasheva

Computers · 15.04.2026 · 15 (4) · 244

DOI: 10.3390/computers15040244 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computers, online erschienen am 15.04.2026. Verfasst von Zhainagul Khamitova und 9 weiteren Autor:innen. Die Arbeit nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Risk stratification of impaired glycemic control remains a major challenge in biomedical data analysis due to heterogeneous metabolic, behavioral, and therapeutic factors observed in large-scale populations. This study proposes a calibrated and interpretable decision–support framework, termed Calibrated Multi-Task Stacking Ensemble (CMSE), for joint modeling of clinically related glycemic outcomes. The framework integrates demographic variables, lipid profiles, renal and inflammatory biomarkers, dietary and smoking indicators, and therapy-related features within a unified predictive architecture. Robust modeling is ensured through leakage-aware preprocessing, quantile-based Winsorization, out-of-fold stacking, and isotonic calibration of probabilistic outputs. The physiological coherence between short-term and long-term glycemic markers is investigated using an explicit intertask coupling mechanism based on the estimated average glucose (eAG) ratio. Model interpretability is supported using SHAP analysis, mutual information, distance correlation, and feature importance metrics. In the primary medication-free screening configuration, the framework is evaluated on the NHANES 2017–March 2020 dataset, achieving ROC-AUC of 0.865 for diabetes classification and R2 values of 0.385 and 0.366 for plasma glucose and HbA1c prediction, respectively. These results indicate that CMSE provides a reliable and explainable approach for calibrated glycemic risk assessment and clinical decision support.

Zitieren

APA

Khamitova, Z., Omarova, G., Akhmetzhanov, M., Burganova, R., Orynbassar, M., Sabirova, U., Bukatayeva, A., Barakova, A., Jiyanmuratova, G., & Yuldasheva, D. (2026). A Calibrated Multi-Task Ensemble Architecture for Biomedical Risk Prediction. Computers, 15(4), 244. https://doi.org/10.3390/computers15040244

Vancouver

Khamitova Z, Omarova G, Akhmetzhanov M, Burganova R, Orynbassar M, Sabirova U, et al. A Calibrated Multi-Task Ensemble Architecture for Biomedical Risk Prediction. Computers. 2026;15(4):244. doi:10.3390/computers15040244

BibTeX
@article{Khamitova2026,
  author = {Khamitova, Zhainagul and Omarova, Gulmira and Akhmetzhanov, Madi and Burganova, Roza and Orynbassar, Maksym and Sabirova, Umida and Bukatayeva, Almagul and Barakova, Aliya and Jiyanmuratova, Gulnoz and Yuldasheva, Dilchekhra},
  title = {A Calibrated Multi-Task Ensemble Architecture for Biomedical Risk Prediction},
  journal = {Computers},
  year = {2026},
  volume = {15},
  number = {4},
  pages = {244},
  doi = {10.3390/computers15040244}
}
Kostenlos starten →