Enhanced XGBoost Approach for Identification and Classification of Alzheimer Disease

Alzheimer's Disease (AD), a progressive neurodegenerative disorder, affects millions worldwide.Early diagnosis and classification are crucial for improving patient outcomes.Over the past decade, Machine Learning (ML) has advanced in predicting and classifying AD.Recently, XGBoost (XGB), a Gradient-Boosting Decision-Tree (GB-DT) based ML model, has shown superior performance.Hence, this work proposes a novel Enhanced XGBoost (E-XGB) model that integrates a feature selection mechanism designed for statistical data of AD patients, addressing common challenges such as class imbalance and overfitting in existing models.The key innovation of E-XGB lies in its ability to automatically select the most relevant features from a broad set of demographics, medical, and cognitive data, thereby optimizing the model's predictive accuracy and computational efficiency.By utilizing the OASIS dataset, which includes both individuals with and without AD, the E-XGB model outperformed traditional ML models, achieving 98.07% accuracy for AD prediction and 98.47% for AD classification.This study highlights the potential of E-XGB as a highly efficient and accurate tool for assisting medical professionals in diagnosing AD and sets the stage for future advancements in the field of neurodegenerative disease prediction.