HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis system

A progressive neurological disorder, Alzheimer’s disease (AD) predominantly impacts memory and cognitive functions of diseased victims. The disease is usually identified by the buildup of abnormal protein deposits in the brain, resulting in the development of plaques and tangles that interfere with...

Full description

Saved in:
Bibliographic Details
Main Authors: S. Veluchamy, R. Bhuvaneswari, K. Ashwini, Samah Alshathri, Walid El-Shafai
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:Engineering Science and Technology, an International Journal
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2215098625001132
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849715009957920768
author S. Veluchamy
R. Bhuvaneswari
K. Ashwini
Samah Alshathri
Walid El-Shafai
author_facet S. Veluchamy
R. Bhuvaneswari
K. Ashwini
Samah Alshathri
Walid El-Shafai
author_sort S. Veluchamy
collection DOAJ
description A progressive neurological disorder, Alzheimer’s disease (AD) predominantly impacts memory and cognitive functions of diseased victims. The disease is usually identified by the buildup of abnormal protein deposits in the brain, resulting in the development of plaques and tangles that interfere with communication between nerve cells. Over time, those affected by Alzheimer’s undergo a diminishing mental capacity, affecting their daily activities and ultimately resulting in a loss of independence. While there is currently no remedy to reverse the advancement of Alzheimer’s disease, identifying the initiation of AD can prove highly beneficial within the medical field. This study presents an innovative framework for the detection and diagnosis of Alzheimer’s disease, employing deep transfer models and COATI optimization techniques. The proposed hybrid Deepnet (HY-Deepnet) framework consists of two main phases: detection and diagnosis. The detection phase aims at identifying the presence of disease with the aid of MRI images. This phase evaluates the performance of various deep learning models, including AlexNet, GoogLenet, SqueezeNet, VGGNet, ResNet and explores their combinations. Experimental results reveal that the combination of AlexNet, GoogLenet, and VGGNet outperforms other networks and their combinations, achieving an accuracy of 77.03% without optimization. The second phase focuses on the diagnosis of the detected disease into three different stages. The diagnosis phase is improved from COATI optimization techniques. The proposed HY-Deepnet thus attains an impressive overall accuracy of 97.6%, accompanied by precision, recall, and F1 scores of 0.978, 0.976, and 0.974, respectively. These results underscore the effectiveness of the framework in enhancing Alzheimer’s detection and diagnosis, particularly when leveraging the synergy of deep transfer models and COATI optimization techniques.
format Article
id doaj-art-8325ff489d224518b70d8ec35ba3f8ea
institution DOAJ
issn 2215-0986
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series Engineering Science and Technology, an International Journal
spelling doaj-art-8325ff489d224518b70d8ec35ba3f8ea2025-08-20T03:13:32ZengElsevierEngineering Science and Technology, an International Journal2215-09862025-08-016810205810.1016/j.jestch.2025.102058HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis systemS. Veluchamy0R. Bhuvaneswari1K. Ashwini2Samah Alshathri3Walid El-Shafai4Department of Electronics and Communication Engineering, Amrita School of Computing, Amrita Vishwa Vidyapeetham, Nagercoil, IndiaDepartment of Computer Science and Engineering, Rajalakshmi Engineering College, Thandalam, Chennai, India; Corresponding author at: Rajalakshmi Engineering College, Thandalam, Chennai, India.Department of Computer Science and Engineering, Amrita School of Computing, Amrita Vishwa Vidyapeetham, Chennai, IndiaDepartment of Information Technology, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh 11671, Saudi ArabiaAutomated Systems and Computing Lab (ASCL), Computer Science Department, Prince Sultan University, Riyadh 11586, Saudi Arabia; Department of Electronics and Electrical Communications Engineering, Faculty of Electronic Engineering, Menoufia University, Menouf 32952, EgyptA progressive neurological disorder, Alzheimer’s disease (AD) predominantly impacts memory and cognitive functions of diseased victims. The disease is usually identified by the buildup of abnormal protein deposits in the brain, resulting in the development of plaques and tangles that interfere with communication between nerve cells. Over time, those affected by Alzheimer’s undergo a diminishing mental capacity, affecting their daily activities and ultimately resulting in a loss of independence. While there is currently no remedy to reverse the advancement of Alzheimer’s disease, identifying the initiation of AD can prove highly beneficial within the medical field. This study presents an innovative framework for the detection and diagnosis of Alzheimer’s disease, employing deep transfer models and COATI optimization techniques. The proposed hybrid Deepnet (HY-Deepnet) framework consists of two main phases: detection and diagnosis. The detection phase aims at identifying the presence of disease with the aid of MRI images. This phase evaluates the performance of various deep learning models, including AlexNet, GoogLenet, SqueezeNet, VGGNet, ResNet and explores their combinations. Experimental results reveal that the combination of AlexNet, GoogLenet, and VGGNet outperforms other networks and their combinations, achieving an accuracy of 77.03% without optimization. The second phase focuses on the diagnosis of the detected disease into three different stages. The diagnosis phase is improved from COATI optimization techniques. The proposed HY-Deepnet thus attains an impressive overall accuracy of 97.6%, accompanied by precision, recall, and F1 scores of 0.978, 0.976, and 0.974, respectively. These results underscore the effectiveness of the framework in enhancing Alzheimer’s detection and diagnosis, particularly when leveraging the synergy of deep transfer models and COATI optimization techniques.http://www.sciencedirect.com/science/article/pii/S2215098625001132COATI optimizationPre-trained networksAlzheimerDeep learningFeature extraction
spellingShingle S. Veluchamy
R. Bhuvaneswari
K. Ashwini
Samah Alshathri
Walid El-Shafai
HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis system
Engineering Science and Technology, an International Journal
COATI optimization
Pre-trained networks
Alzheimer
Deep learning
Feature extraction
title HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis system
title_full HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis system
title_fullStr HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis system
title_full_unstemmed HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis system
title_short HY-Deepnet : A new Optimal Deep transfer learning empowered framework for an autonomous Alzheimer’s disease detection and diagnosis system
title_sort hy deepnet a new optimal deep transfer learning empowered framework for an autonomous alzheimer s disease detection and diagnosis system
topic COATI optimization
Pre-trained networks
Alzheimer
Deep learning
Feature extraction
url http://www.sciencedirect.com/science/article/pii/S2215098625001132
work_keys_str_mv AT sveluchamy hydeepnetanewoptimaldeeptransferlearningempoweredframeworkforanautonomousalzheimersdiseasedetectionanddiagnosissystem
AT rbhuvaneswari hydeepnetanewoptimaldeeptransferlearningempoweredframeworkforanautonomousalzheimersdiseasedetectionanddiagnosissystem
AT kashwini hydeepnetanewoptimaldeeptransferlearningempoweredframeworkforanautonomousalzheimersdiseasedetectionanddiagnosissystem
AT samahalshathri hydeepnetanewoptimaldeeptransferlearningempoweredframeworkforanautonomousalzheimersdiseasedetectionanddiagnosissystem
AT walidelshafai hydeepnetanewoptimaldeeptransferlearningempoweredframeworkforanautonomousalzheimersdiseasedetectionanddiagnosissystem