AI ensemble predicts early Alzheimer's with 97% accuracy, identifies key biomarkers
A stacking ensemble of 4 ML models plus an ANN outperforms single classifiers on ADNI data.
Deep Dive
Researcher Debopriya Ghosh developed a stacking ensemble model (Logistic Regression, Extra Trees, Bagging KNN, LightGBM) and an artificial neural network to detect early-stage Alzheimer's using clinical, neuropsychological, and neuroimaging data from ADNI. The study applied iterative imputation for missing values and Borderline SVM-SMOTE for class imbalance. Feature selection reduced dimensionality. The performance of these models is compared using precision, recall, F1-score, and AUC-ROC. The study aims to find the
Key Points
- Stacking ensemble uses Logistic Regression, Extra Trees, Bagging KNN, and LightGBM as base classifiers.
- Data preprocessing includes iterative imputation for missing values and Borderline SVM-SMOTE for class imbalance.
- Models identify key biomarkers from clinical, neuropsychological, and neuroimaging data for early Alzheimer's detection.
Why It Matters
Early Alzheimer's detection enables timely intervention; this model offers a powerful, data-driven tool for clinicians.