Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Transfer Learning-Assisted DementiaNet: A Four Layer Deep CNN for Accurate Alzheimer’s Disease Detection from MRI Images
Khulna University of Engineering & Technology, Khulna, Bangladesh.ORCID iD: 0000-0002-0854-6732
Department of Computer Science and Engineering, Green University of Bangladesh, Dhaka, Bangladesh.ORCID iD: 0000-0002-3207-9855
Khulna University of Engineering & Technology, Khulna, Bangladesh.ORCID iD: 0000-0002-7341-7574
Department of Computer Science and Engineering, International Islamic University Chittagong, Chittagong, Bangladesh.
Show others and affiliations
2023 (English)In: BI 2023: Brain Informatics, Proceedings / [ed] Yu Zhang, Hongzhi Kuai & Emily P. Stephen, Springer, 2023, p. 383-394Conference paper, Published paper (Refereed)
Abstract [en]

Alzheimer’s disease is the most common type of dementia and the sixth highest cause of mortality among people over 65. Also, according to statistics, the number of deaths from Alzheimer’s disease has increased dramatically. As a result, early detection of Alzheimer’s disease can improve patient survival chances. Machine learning methods using magnetic resonance imaging have been utilized in the diagnosis of Alzheimer’s disease to help clinicians and speed up the procedure. This research mainly focuses on Alzheimer’s disease detection to overcome previous limitations. We use a publicly available dataset which contains 6400 MRI images. In order to train our dataset, we employ our suggested model, “DementiaNet”, using “EfficientNet” as a feature extractor and a Deep CNN as a classifier. In order to capture all the features, this framework uses a small number of convolutional layers, which improves the effectiveness of feature learning and results in a more accurate and reliable output. In addition, to address the issue of data imbalance, we apply data augmentation to enhance the size of the minority class by considering four stages of dementia. Also, in this study, we take advantage of the transfer learning approach with the attachment of “EfficientNet”, which allows our model to easily solve the overfitting problem and also extract all the features in a effective way. Finally, we compare our proposed models with those reported in the literature. Our experimental results indicate that our “DementiaNet” model achieve an overall classification accuracy of 97% to detect Alzheimer from brain MRI images, exceeding all other state-of-the-art models.

Place, publisher, year, edition, pages
Springer, 2023. p. 383-394
Series
Lecture Notes in Artificial Intelligence (Lecture Notes in Computer Science), ISSN 0302-9743, E-ISSN 1611-3349 ; 13974
Keywords [en]
Alzheimer’s disease, Convolutional Neural Network, Data Augmentation, DementiaNet, EfficientNet, Feature Extraction, MRI, Preprocessing, Transfer Learning
National Category
Medical Imaging Computer Sciences
Research subject
Pervasive Mobile Computing; Cyber Security
Identifiers
URN: urn:nbn:se:ltu:diva-103371DOI: 10.1007/978-3-031-43075-6_33Scopus ID: 2-s2.0-85172413214OAI: oai:DiVA.org:ltu-103371DiVA, id: diva2:1823758
Conference
16th International Conference on Brain Informatics (BI 2023), Hoboken, NJ, United States, August 1-3, 2023
Note

ISBN for host publication: 978-3-031-43074-9  (print), 978-3-031-43075-6 (electronic)

Available from: 2024-01-03 Created: 2024-01-03 Last updated: 2025-02-09Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Hossain, Mohammad ShahadatAndersson, Karl

Search in DiVA

By author/editor
Umme Habiba, SultanaDebnath, TanoyIslam, Md. KhairulHossain, Mohammad ShahadatAndersson, Karl
By organisation
Computer Science
Medical ImagingComputer Sciences

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 41 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf