Deep Learning based Framework for Early-stage Alzheimer Detection
Procedia Computer Science · 2026 · 283 · 3063-3073
DOI: 10.1016/j.procs.2026.06.372 ↗Kurz gefasst
Fachartikel in der Zeitschrift Procedia Computer Science, 2026. Verfasst von Venkata Seshaiah Banka und 3 weiteren Autor:innen. Die Arbeit nennt 9 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Themen: Demenz
Abstract (Originalsprache)
Alzheimer’s Disease (AD) is one of the most important global health challenges, and early detection is one of the most important tools for successful treatment. In this article, develop a novel system for cognitive appraisal through brain imaging features for AD detection. Using Convolutional Neural Networks (CNNs), two models were trained separately and classified with a validation accuracy of 98%. The dataset has four class labels with ‘Mild Demented’, ‘Moderate Demented’, ‘Non-Demented’ and ‘Very Mild Demented’. In this study, basic CNN model predictions were consolidated using Alzheimer’s detection based on deep Learning. The proposed ensemble model outperformed individual CNN demonstrating improved accuracy and robustness of AD diagnosis. we got good results on standard Metrics such as F1-score (98%), recall (98%), accuracy (98%), and precision (97%) showcased the efficiency of the ensemble technique for early detection of AD.
Zitieren
APA
Banka, V. S., Thota, S. M., Chandra, N. C., & Dara, S. (2026). Deep Learning based Framework for Early-stage Alzheimer Detection. Procedia Computer Science, 283, 3063-3073. https://doi.org/10.1016/j.procs.2026.06.372
Vancouver
Banka VS, Thota SM, Chandra NC, Dara S. Deep Learning based Framework for Early-stage Alzheimer Detection. Procedia Computer Science. 2026;283:3063-3073. doi:10.1016/j.procs.2026.06.372
BibTeX
@article{Banka2026,
author = {Banka, Venkata Seshaiah and Thota, Sai Mokshitha and Chandra, Neha Chowdary and Dara, Suresh},
title = {Deep Learning based Framework for Early-stage Alzheimer Detection},
journal = {Procedia Computer Science},
year = {2026},
volume = {283},
pages = {3063--3073},
doi = {10.1016/j.procs.2026.06.372}
}
EUVIMED