A dual-model AI framework for Alzheimer’s disease diagnosis using clinical and MRI data

TL;DR

This paper presents a new AI framework combining clinical data and MRI scans to improve Alzheimer's disease diagnosis accuracy and early detection.

Contribution

A novel dual-model AI framework that integrates clinical and MRI data for more accurate Alzheimer’s disease diagnosis.

Findings

The ANN achieved 87.08% accuracy in early-stage Alzheimer’s risk prediction using clinical data.

The CNN demonstrated 97% accuracy in disease staging using MRI images with Grad-CAM visualizations.

The dual-model approach improves diagnostic precision and interpretability compared to traditional methods.

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that requires advanced diagnostic strategies for early and accurate detection.

This study introduces a hybrid AI-driven diagnostic framework that integrates an Artificial Neural Network (ANN) trained on clinical data from 1,200 patients using 31 demographic, symptomatic, and behavioral features with a Convolutional Neural Network (CNN) trained on 4,876 MRI images to classify AD into four stages.

The ANN achieved an accuracy of 87.08% in early-stage risk prediction, while the CNN demonstrated a superior 97% accuracy in disease staging, supported by Grad-CAM visualizations that improved model interpretability. This dual-model approach effectively combines structured clinical data with imaging-based analysis, addressing the sensitivity and scalability limitations of traditional diagnostic methods and providing a more comprehensive assessment of AD.

The integration of ANN and CNN enhances diagnostic precision and supports AI-assisted clinical decision-making, with future work focusing on lightweight CNN architectures and wearable technologies to enable broader accessibility and earlier intervention.

Figures