Home / Current Issue / Paper 1720288
A Hybrid Kernel SHAP and Enhanced ALE Framework for Explainable Stroke Risk Assessment
Subject area: Physical Sciences and Environment · Area of research: Machine Learning, Health
DOI: https://doi.org/10.64388/IREV10I2-1720288
Abstract
Stroke remains a leading cause of death and disability worldwide, yet machine learning (ML) models capable of predicting risk are rarely trusted clinically due to their "black-box" nature. This study introduces an explainability-centered hybrid framework, ALE-SHAP, evaluated on 5,110 patient records. A class-weighted Random Forest classifier addressed severe class imbalance (4.9% stroke prevalence), achieving 86.89%accuracy and an AUC-ROC of 0.783. To open the black box, we paired Kernel SHAP which identified age and average glucose level as dominant risk drivers with a novel Enhanced Accumulated Local Effects (ALE) algorithm featuring adaptive binning, spline smoothing, and bootstrapped confidence intervals. Unlike traditional Partial Dependence Plots, which introduced extrapolation errors due to feature multi-collinearity (r=0.33 between age and BMI), our Enhanced ALE revealed clinically valid, non-linear risk thresholds: a sharp inflection point at age 50, an inverted U-shaped BMI curve peaking at 28.4〖" kg/m" 〗^2(mirroring the clinical "obesity paradox"), and a dual-peak glucose risk profile. The framework achieved high explanation fidelity (D=0.925) and stability (S=0.884), offering a reliable, clinically aligned tool for early stroke prevention.
Keywords
Accumulated Local Effects, Explainable Artificial Intelligence, Kernel SHAP, Machine Learning, Stroke Prediction.
How to cite this paper
@article{1720288,
author = {Okwueze Chisom Nneoma, MBA Chioma Juliet, Agbo Chika},
title = {A Hybrid Kernel SHAP and Enhanced ALE Framework for Explainable Stroke Risk Assessment},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {10},
number = {2},
pages = {133-139},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1720288.pdf},
abstract = {Stroke remains a leading cause of death and disability worldwide, yet machine learning (ML) models capable of predicting risk are rarely trusted clinically due to their "black-box" nature. This study introduces an explainability-centered hybrid framework, ALE-SHAP, evaluated on 5,110 patient records. A class-weighted Random Forest classifier addressed severe class imbalance (4.9% stroke prevalence), achieving 86.89%accuracy and an AUC-ROC of 0.783. To open the black box, we paired Kernel SHAP which identified age and average glucose level as dominant risk drivers with a novel Enhanced Accumulated Local Effects (ALE) algorithm featuring adaptive binning, spline smoothing, and bootstrapped confidence intervals. Unlike traditional Partial Dependence Plots, which introduced extrapolation errors due to feature multi-collinearity (r=0.33 between age and BMI), our Enhanced ALE revealed clinically valid, non-linear risk thresholds: a sharp inflection point at age 50, an inverted U-shaped BMI curve peaking at 28.4〖" kg/m" 〗^2(mirroring the clinical "obesity paradox"), and a dual-peak glucose risk profile. The framework achieved high explanation fidelity (D=0.925) and stability (S=0.884), offering a reliable, clinically aligned tool for early stroke prevention.},
keywords = {Accumulated Local Effects, Explainable Artificial Intelligence, Kernel SHAP, Machine Learning, Stroke Prediction.},
month = {August},
doi = {https://doi.org/10.64388/IREV10I2-1720288}
}