What question did this study set out to answer?

To propose a hybrid Explainable AI framework that enhances fair automated decision-making in healthcare.

March 8, 2026Open Access

A Hybrid Explainable AI Framework for Fair Automated Decision-Making in Healthcare

View Full Paper

Ask AI

Bookmark

Authors

Proposes a hybrid AI framework for improving fairness and interpretability in healthcare decision-making systems.

To propose a hybrid Explainable AI framework that enhances fair automated decision-making in healthcare.
Constructed a dual pipeline processing metadata and chest radiographs for binary classification.
Employed Logistic Regression, Random Forest, and Convolutional Neural Network models.
Managed class imbalance using oversampling and SMOTE.
Utilized SHAP and LIME for explainability and Grad-CAM for visualization.
Quantified fairness using demographic metrics and mitigated disparities with optimization techniques.
Balancing strategies improved Effusion recall in both pipelines.
Hybrid explanations provided insights supporting clinical reasoning.
Fairness analysis highlighted measurable disparities which were reduced without major performance loss.

Urooj et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark