Multi-model XAI framework for prediabetes and diabetes risk stratification using population-level health indicators
Introduction Diabetes mellitus is a globally prevalent chronic metabolic disorder whose timely identification is critical to preventing irreversible end-organ damage. Although machine learning (ML) approaches have shown considerable promise for automated diabetes screening, the field is hampered by methodological deficiencies — notably data leakage during preprocessing and oversampler selection performed outside cross-validation folds — that collectively produce overoptimistic and clinically unreliable performance estimates. Methods We present a rigorously leakage-aware, multi-stage ML pipeline evaluated on a large-scale publicly available synthetic clinical dataset comprising 100,000 records and 15 features (prevalence 8.5%). The pipeline explicitly addresses two categories of leakage: (i) preprocessing leakage , by fitting all transformers exclusively on training data; and (ii) resampling leakage , by performing oversampler comparison entirely within cross-validation folds. The pipeline encompasses: (i) stratified three-way splitting into training, calibration, and held-out test partitions; (ii) an in-fold cross-validated comparison of seven resampling strategies against a no-resampling class-weighted baseline, with SMOTETomek selected automatically on cross-validated F1; (iii) training of four base classifiers (Logistic Regression, Random Forest, HistGradientBoosting, LightGBM); (iv) six model-fusion strategies; (v) isotonic probability calibration; (vi) decision curve analysis; and (vii) Mondrian class-conditional split-conformal prediction. Results Among all models, LightGBM, HistGradientBoosting, and the logarithmic pooling fusion ensemble (Fus-LogPool) achieved statistically equivalent top performance: PR-AUC of 0.8807 (Fus-LogPool), 0.8801 (LightGBM), and 0.8798 (HistGB), with pairwise McNemar tests uniformly non-significant ( p > 0.5), indicating that a simple well-tuned single gradient-boosted tree is as effective as fusion on this dataset. Mondrian conformal prediction attained 90.39% empirical coverage (target 90%), with 98.82% singleton predictions and singleton accuracy of 91.46%. Decision curve analysis confirmed net clinical benefit over treat-all and treat-none strategies across clinically relevant risk thresholds (5%–20%). SHAP global explanations identified HbA1c level, blood glucose level, BMI, and age as the dominant risk determinants. Discussion Importantly, all findings derive from a synthetic or semi-synthetic dataset without a verified clinical source, and must be validated on prospective real-world cohorts before any clinical application. The fully reproducible pipeline is released as an open-source Kaggle notebook.
Authors
- Fahim Abrar
- Ibtissam Essadik
- Dibyodyuti Samaddar
- Mahmood Ghaleb Al-Bashayreh (ORCID: https://orcid.org/0000-0002-8583-0673)
- Latifa Mochhoury (ORCID: https://orcid.org/0000-0002-5019-2296)
- Md Tanzin Rahman
- Shima Ali Sadia (ORCID: https://orcid.org/0009-0006-1086-607X)
- Anandhavalli Muniasamy
- Zarafruz Burkhonova
Institutions
- University of Technology Sydney (AU)
- Applied Science Private University (JO)
- Monroe University (US)
- Trine University (US)
- École Normale Supérieure - Marrakech (MA)
- Samarkand State Medical Institute (UZ)
- Université Hassan 1er (MA)
- Sup de Co Marrakech (MA)
- Université Privée de Marrakech (MA)
- Institute of Medical Sciences (IN)
- King Khalid University (SA)
- College of New Rochelle (US)
- Banaras Hindu University (IN)
Publication Details
- Journal
- Frontiers in Medicine
- Published
- 2026-11-17
- DOI
- https://doi.org/10.3389/fmed.2026.1986834
- Primary Topic
- Artificial Intelligence in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00