Automated Diagnosis of Knee Osteoarthritis: A Stacked Ensemble Deep Learning Approach With Explainable AI Techniques

Knee osteoarthritis (KOA) is a widespread degenerative joint disease that poses significant global challenges owing to its delayed diagnosis and treatment, often resulting in severe disability. Despite extensive research on predicting KOA, many of the proposed methods lack reliability because they fail to incorporate explainable AI (XAI) methodologies, robust preprocessing techniques, and appropriate hyperparameter tuning. This study introduces a deep learning framework for KOA classification, addressing both binary (diagnosis) and multi-class (severity prediction) classification tasks using the Osteoarthritis Initiative (OAI) dataset. Our approach employs a comprehensive image preprocessing pipeline. This pipeline includes scaling, sharpening, denoising, histogram equalization, and contrast enhancement to standardize image quality and highlight crucial features for classification. The proposed stacked ensemble model, which integrates Xception, EfficientNetB5, and InceptionV3, surpasses individual models, achieving 86.29% accuracy in KOA diagnosis and 96.93% in KOA severity prediction. To ensure transparency and interpretability, we incorporated advanced explainability tools, including Gradient-weighted Class Activation Mapping (Grad-CAM), Faster Score-CAM, and Local Interpretable Model-agnostic Explanations (LIME). These tools provide clear visual insights into the model’s decision-making process. Our findings present a balanced approach that combines performance with transparency, potentially leading to earlier and more accurate KOA diagnoses.