Enhancing Alzheimer’s Disease Detection: A Hybrid Approach with CNNs and Vision Transformers

Vimbi Viswan; Noushath Shaffi; Srinivas Rao Sirasanagandla; Mufti Mahmud

doi:10.1007/978-3-032-04657-4_18

Enhancing Alzheimer’s Disease Detection: A Hybrid Approach with CNNs and Vision Transformers

Vimbi Viswan
, Noushath Shaffi
, Srinivas Rao Sirasanagandla
, Mufti Mahmud^*

^*Corresponding author for this work

Department of Information and Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Alzheimer’s disease (AD) leads to memory loss and cognitive issues, making early detection and treatment essential. This study explores a new approach to identifying AD using medical imaging techniques. It focuses on combining two types of image analysis methods: Convolutional Neural Networks (CNNs) and Vision Transformers (VT). CNNs are well-known for analyzing images by identifying local patterns and details effectively. However, VT models excel at capturing the overall structure and relationships in an image, providing a broader perspective. Inspired by the strengths of both methods, this research introduces a combined approach that integrates CNN and VT to create a more reliable system for detecting Alzheimer’s disease. The hybrid model was tested using data from two well-known Alzheimer’s disease datasets: OASIS and ADNI. Its performance was then compared with leading individual CNN and VT models under various testing conditions. The results showed that this hybrid method outperformed the separate models. It achieved an impressive accuracy rate of 98.5% on one dataset and 99.6% on the other, demonstrating its effectiveness and reliability in diagnosing AD. This study highlights the potential of combining different image analysis techniques to improve medical imaging applications, offering a promising tool for early detection of Alzheimer’s disease.

Original language	English
Title of host publication	Applied Intelligence and Informatics - 4th International Conference, AII 2024, Revised Selected Papers
Editors	Mufti Mahmud, M. Shamim Kaiser, Joarder Kamruzzaman, Khan Iftekharuddin, Md Atiqur Rahman Ahad, Ning Zhong
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	264-276
Number of pages	13
ISBN (Print)	9783032046567
DOIs	https://doi.org/10.1007/978-3-032-04657-4_18
State	Published - 2025
Event	4th International Conference on Applied Intelligence and Informatics, AII 2024 - London, United Kingdom Duration: 18 Dec 2024 → 20 Dec 2024

Publication series

Name	Communications in Computer and Information Science
Volume	2607 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	4th International Conference on Applied Intelligence and Informatics, AII 2024
Country/Territory	United Kingdom
City	London
Period	18/12/24 → 20/12/24

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Keywords

Alzheimer’s Disease
Convolutional Neural Networks
Hybrid architecture
Machine Learning models
Vision Transformers

ASJC Scopus subject areas

General Computer Science
General Mathematics

Access to Document

10.1007/978-3-032-04657-4_18

Cite this

Viswan, V., Shaffi, N., Sirasanagandla, S. R., & Mahmud, M. (2025). Enhancing Alzheimer’s Disease Detection: A Hybrid Approach with CNNs and Vision Transformers. In M. Mahmud, M. S. Kaiser, J. Kamruzzaman, K. Iftekharuddin, M. A. R. Ahad, & N. Zhong (Eds.), Applied Intelligence and Informatics - 4th International Conference, AII 2024, Revised Selected Papers (pp. 264-276). (Communications in Computer and Information Science; Vol. 2607 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-032-04657-4_18

Viswan, Vimbi ; Shaffi, Noushath ; Sirasanagandla, Srinivas Rao et al. / Enhancing Alzheimer’s Disease Detection : A Hybrid Approach with CNNs and Vision Transformers. Applied Intelligence and Informatics - 4th International Conference, AII 2024, Revised Selected Papers. editor / Mufti Mahmud ; M. Shamim Kaiser ; Joarder Kamruzzaman ; Khan Iftekharuddin ; Md Atiqur Rahman Ahad ; Ning Zhong. Springer Science and Business Media Deutschland GmbH, 2025. pp. 264-276 (Communications in Computer and Information Science).

@inproceedings{969c5cb446864e8882016ae21ecaa9b8,

title = "Enhancing Alzheimer{\textquoteright}s Disease Detection: A Hybrid Approach with CNNs and Vision Transformers",

abstract = "Alzheimer{\textquoteright}s disease (AD) leads to memory loss and cognitive issues, making early detection and treatment essential. This study explores a new approach to identifying AD using medical imaging techniques. It focuses on combining two types of image analysis methods: Convolutional Neural Networks (CNNs) and Vision Transformers (VT). CNNs are well-known for analyzing images by identifying local patterns and details effectively. However, VT models excel at capturing the overall structure and relationships in an image, providing a broader perspective. Inspired by the strengths of both methods, this research introduces a combined approach that integrates CNN and VT to create a more reliable system for detecting Alzheimer{\textquoteright}s disease. The hybrid model was tested using data from two well-known Alzheimer{\textquoteright}s disease datasets: OASIS and ADNI. Its performance was then compared with leading individual CNN and VT models under various testing conditions. The results showed that this hybrid method outperformed the separate models. It achieved an impressive accuracy rate of 98.5\% on one dataset and 99.6\% on the other, demonstrating its effectiveness and reliability in diagnosing AD. This study highlights the potential of combining different image analysis techniques to improve medical imaging applications, offering a promising tool for early detection of Alzheimer{\textquoteright}s disease.",

keywords = "Alzheimer{\textquoteright}s Disease, Convolutional Neural Networks, Hybrid architecture, Machine Learning models, Vision Transformers",

author = "Vimbi Viswan and Noushath Shaffi and Sirasanagandla, \{Srinivas Rao\} and Mufti Mahmud",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.; 4th International Conference on Applied Intelligence and Informatics, AII 2024 ; Conference date: 18-12-2024 Through 20-12-2024",

year = "2025",

doi = "10.1007/978-3-032-04657-4\_18",

language = "English",

isbn = "9783032046567",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "264--276",

editor = "Mufti Mahmud and Kaiser, \{M. Shamim\} and Joarder Kamruzzaman and Khan Iftekharuddin and Ahad, \{Md Atiqur Rahman\} and Ning Zhong",

booktitle = "Applied Intelligence and Informatics - 4th International Conference, AII 2024, Revised Selected Papers",

address = "Germany",

}

Viswan, V, Shaffi, N, Sirasanagandla, SR & Mahmud, M 2025, Enhancing Alzheimer’s Disease Detection: A Hybrid Approach with CNNs and Vision Transformers. in M Mahmud, MS Kaiser, J Kamruzzaman, K Iftekharuddin, MAR Ahad & N Zhong (eds), Applied Intelligence and Informatics - 4th International Conference, AII 2024, Revised Selected Papers. Communications in Computer and Information Science, vol. 2607 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 264-276, 4th International Conference on Applied Intelligence and Informatics, AII 2024, London, United Kingdom, 18/12/24. https://doi.org/10.1007/978-3-032-04657-4_18

Enhancing Alzheimer’s Disease Detection: A Hybrid Approach with CNNs and Vision Transformers. / Viswan, Vimbi; Shaffi, Noushath; Sirasanagandla, Srinivas Rao et al.
Applied Intelligence and Informatics - 4th International Conference, AII 2024, Revised Selected Papers. ed. / Mufti Mahmud; M. Shamim Kaiser; Joarder Kamruzzaman; Khan Iftekharuddin; Md Atiqur Rahman Ahad; Ning Zhong. Springer Science and Business Media Deutschland GmbH, 2025. p. 264-276 (Communications in Computer and Information Science; Vol. 2607 CCIS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Enhancing Alzheimer’s Disease Detection

T2 - 4th International Conference on Applied Intelligence and Informatics, AII 2024

AU - Viswan, Vimbi

AU - Shaffi, Noushath

AU - Sirasanagandla, Srinivas Rao

AU - Mahmud, Mufti

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

PY - 2025

Y1 - 2025

N2 - Alzheimer’s disease (AD) leads to memory loss and cognitive issues, making early detection and treatment essential. This study explores a new approach to identifying AD using medical imaging techniques. It focuses on combining two types of image analysis methods: Convolutional Neural Networks (CNNs) and Vision Transformers (VT). CNNs are well-known for analyzing images by identifying local patterns and details effectively. However, VT models excel at capturing the overall structure and relationships in an image, providing a broader perspective. Inspired by the strengths of both methods, this research introduces a combined approach that integrates CNN and VT to create a more reliable system for detecting Alzheimer’s disease. The hybrid model was tested using data from two well-known Alzheimer’s disease datasets: OASIS and ADNI. Its performance was then compared with leading individual CNN and VT models under various testing conditions. The results showed that this hybrid method outperformed the separate models. It achieved an impressive accuracy rate of 98.5% on one dataset and 99.6% on the other, demonstrating its effectiveness and reliability in diagnosing AD. This study highlights the potential of combining different image analysis techniques to improve medical imaging applications, offering a promising tool for early detection of Alzheimer’s disease.

AB - Alzheimer’s disease (AD) leads to memory loss and cognitive issues, making early detection and treatment essential. This study explores a new approach to identifying AD using medical imaging techniques. It focuses on combining two types of image analysis methods: Convolutional Neural Networks (CNNs) and Vision Transformers (VT). CNNs are well-known for analyzing images by identifying local patterns and details effectively. However, VT models excel at capturing the overall structure and relationships in an image, providing a broader perspective. Inspired by the strengths of both methods, this research introduces a combined approach that integrates CNN and VT to create a more reliable system for detecting Alzheimer’s disease. The hybrid model was tested using data from two well-known Alzheimer’s disease datasets: OASIS and ADNI. Its performance was then compared with leading individual CNN and VT models under various testing conditions. The results showed that this hybrid method outperformed the separate models. It achieved an impressive accuracy rate of 98.5% on one dataset and 99.6% on the other, demonstrating its effectiveness and reliability in diagnosing AD. This study highlights the potential of combining different image analysis techniques to improve medical imaging applications, offering a promising tool for early detection of Alzheimer’s disease.

KW - Alzheimer’s Disease

KW - Convolutional Neural Networks

KW - Hybrid architecture

KW - Machine Learning models

KW - Vision Transformers

UR - https://www.scopus.com/pages/publications/105020239609

U2 - 10.1007/978-3-032-04657-4_18

DO - 10.1007/978-3-032-04657-4_18

M3 - Conference contribution

AN - SCOPUS:105020239609

SN - 9783032046567

T3 - Communications in Computer and Information Science

SP - 264

EP - 276

BT - Applied Intelligence and Informatics - 4th International Conference, AII 2024, Revised Selected Papers

A2 - Mahmud, Mufti

A2 - Kaiser, M. Shamim

A2 - Kamruzzaman, Joarder

A2 - Iftekharuddin, Khan

A2 - Ahad, Md Atiqur Rahman

A2 - Zhong, Ning

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 18 December 2024 through 20 December 2024

ER -

Viswan V, Shaffi N, Sirasanagandla SR, Mahmud M. Enhancing Alzheimer’s Disease Detection: A Hybrid Approach with CNNs and Vision Transformers. In Mahmud M, Kaiser MS, Kamruzzaman J, Iftekharuddin K, Ahad MAR, Zhong N, editors, Applied Intelligence and Informatics - 4th International Conference, AII 2024, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2025. p. 264-276. (Communications in Computer and Information Science). doi: 10.1007/978-3-032-04657-4_18