Alzheimer’s Disease Detection Using Lightweight Convolutional Neural Network on MRI Scans
American Journal of Bioscience and Bioinformatics · 2025
DOI: 10.54536/ajbb.v4i1.5943 ↗Kurz gefasst
Fachartikel in der Zeitschrift American Journal of Bioscience and Bioinformatics, 2025. Verfasst von Venant Niyonkuru und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Themen: Demenz
Abstract (Originalsprache)
This research suggests a lightweight Convolutional Neural Network (CNN) model for Alzheimer’s Disease (AD) detection and classification from Magnetic Resonance Imaging (MRI) scans. Two models were trained: a binary model distinguishing AD from non-AD cases and a multi-class model distinguishing four stages of dementia (Non-Demented, Very Mild-Demented, Mild Demented, and Moderate Demented). The models were trained on a Kaggle dataset with high accuracy (97.5% binary, 89.3% multi-class). The research demonstrates the viability of lightweight CNNs for low-cost and precise AD diagnosis, with class imbalance managed through data augmentation.
Zitieren
APA
Niyonkuru, V., Sylla, S., Kasuli, Z. A., Sinzinkayo, J. J., & Kabanga, D. (2025). Alzheimer’s Disease Detection Using Lightweight Convolutional Neural Network on MRI Scans. American Journal of Bioscience and Bioinformatics. https://doi.org/10.54536/ajbb.v4i1.5943
Vancouver
Niyonkuru V, Sylla S, Kasuli ZA, Sinzinkayo JJ, Kabanga D. Alzheimer’s Disease Detection Using Lightweight Convolutional Neural Network on MRI Scans. American Journal of Bioscience and Bioinformatics. 2025. doi:10.54536/ajbb.v4i1.5943
BibTeX
@article{Niyonkuru2025,
author = {Niyonkuru, Venant and Sylla, Sekou and Kasuli, Zindazed Abdshahd and Sinzinkayo, Jimmy Jackson and Kabanga, Deo},
title = {Alzheimer’s Disease Detection Using Lightweight Convolutional Neural Network on MRI Scans},
journal = {American Journal of Bioscience and Bioinformatics},
year = {2025},
doi = {10.54536/ajbb.v4i1.5943}
}
EUVIMED