AlzhEse: A Comprehensive Approach Utilizing CNNs for Alzheimer’s Detection

DOI

10.2991/978-94-6463-762-5_10How to use a DOI?

Keywords

Alzheimer’s Disease Detection; Alzheimer’s Disease Classification; Image Segmentation; Multi-Class Classification; Deep Learning; Convolutional Neural Network (CNN)

Abstract

Convolutional Neural Networks (CNNs) have shown great promise in handling the challenging task of classifying MRI images, particularly in the identification of brain tumors and the diagnosis of Alzheimer’s disease. Finding the ideal values for these parameters can still be difficult because of the complexity of the search space and the possibility of receiving less-than-ideal results, even though CNNs automatically adjust their parameters through training processes. This presents a major challenge to the development of practical applications utilizing CNNs for the processing of MRI images. The model performed very well at four stages: Non-Demented, Very Mild Demented, Mild Demented, and Moderate Demented. Techniques of early stopping, batch normalization, dropout, and data augmentation techniques were applied during training, thus ensuring a balanced performance for all classes with minimum bias and a high reliance in the diagnosis process. These results well establish balanced datasets usage and appropriate preprocessing of images in a hospital environment and point out that the proposed model is capable of early-stage detection and reliable classification for AD. Detection of Alzheimer’s disease is vital for timely healthcare intervention, and in general, it becomes crucial for improving patient outcomes. Deep learning has become an essential tool for detection through the automation of processes. We present a CNN with dense layers model that is efficient for classifying Alzheimer’s disease from MRI images. Categories within the dataset are classified into four types—Non-Demented, Mild Demented, Very Mild Demented, and Moderate Demented. Median filtering and normalization were used to enhance image quality. Our CNN model comprises different convolutional layers and pooling layers followed by fully connected layers with dropout regularization to avoid overfitting. Data augmentation methods were applied to enhance the generalization capability of the model. This model was trained and validated on a balanced dataset with an accuracy of 91.95% on test data, making possible the automation of Alzheimer’s disease detection and minimizing the possibility of human error. Experimental results demonstrated that deep learning models like CNN can sig- nificantly assist in medical diagnostics with efficient and accurate classification of Alzheimer’s disease.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - P. Sanjeev Karthick
AU  - D. M. Rakkesh
AU  - J. Jeyalakshmi
PY  - 2025
DA  - 2025/06/16
TI  - AlzhEse: A Comprehensive Approach Utilizing CNNs for Alzheimer’s Detection
BT  - Proceedings of the International Conference on Materials, Energy, Environment & Manufacturing Sciences & Computational Intelligence and Smart Communication (MEEMS-CISC-2024)
PB  - Atlantis Press
SP  - 100
EP  - 110
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-762-5_10
DO  - 10.2991/978-94-6463-762-5_10
ID  - Karthick2025
ER  -