Advancing Alzheimer’s Diagnosis: A Comparative Analysis of Deep Learning Architectures on Multidimensional Health Data

Alzheimer’s Disease (AD) is a leading cause of disability among the elderly, with its prevalence projected to triple by 2050. Early detection remains critical for effective disease management, yet traditional diagnostic methods are often time-intensive and subjective. This study investigates the effectiveness of three machine learning architectures: Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) in detecting Alzheimer’s Disease using a multidimensional dataset comprising demographic, lifestyle, medical, cognitive, and functional data from 2,149 patients. Each model was evaluated using 10-fold cross-validation, with performance metrics including accuracy, precision, recall, and F1-score. The CNN model demonstrated superior performance, achieving an average accuracy of 88.65%, surpassing both the MLP (84.41%) and LSTM (75.57%) models. These results highlight CNNs’ capability to effectively extract spatial patterns in health data, making them a promising tool for Alzheimer’s diagnosis. In contrast, LSTM underperformed due to the lack of temporal relationships in the dataset. This study underscores the importance of aligning model architecture with dataset characteristics and provides a foundation for integrating machine learning into clinical workflows. Future work will focus on hybrid architectures and real-world validation to enhance diagnostic accuracy and scalability.