Publication detail

Multi-scale Attention Network for Early Detection of Alzheimer’s Disease from MRI images

AGGARWAL, V. SAHU, G. DUTTA, M. JONÁK, M. BURGET, R.

Original Title

Multi-scale Attention Network for Early Detection of Alzheimer’s Disease from MRI images

Type

conference paper

Language

English

Original Abstract

Alzheimer’s disease (AD) is a chronic neurodegenerative disorder that affects brain cells and causes irreversible memory loss, often known as dementia. Many individuals die from this disease each year due to its incurable nature. However, the timely identification of the ailment can play a pivotal role in mitigating its progression. Nowadays, deep learning is used to design an automated system that can detect and classify AD in the early stages. Thus, a novel multi-scale attention network (MSAN-Net) is introduced in this study. The proposed technique uses brain magnetic resonance imaging (MRI) to categorize images into four stages; non-demented, mild demented, very mild demented, and moderate demented. The proposed work is compared with four state-of-the-art methods, and the experimental results suggest that the MSAN-Net exhibits superior performance than the compared approaches.

Keywords

Alzheimer’s disease, deep learning, MRI, multi-class classification, attention network

Authors

AGGARWAL, V.; SAHU, G.; DUTTA, M.; JONÁK, M.; BURGET, R.

Released

30. 10. 2023

Publisher

IEEE

Location

Ghent, Belgium

ISBN

979-8-3503-9328-6

Book

2023 15th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)

Pages from

50

Pages to

55

Pages count

6

URL

https://ieeexplore.ieee.org/document/10333096

BibTex

@inproceedings{BUT185771,
  author="Vaishali {Aggarwal} and Geet {Sahu} and Malay Kishore {Dutta} and Martin {Jonák} and Radim {Burget}",
  title="Multi-scale Attention Network for Early Detection of Alzheimer’s Disease from MRI images",
  booktitle="2023 15th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)",
  year="2023",
  pages="50--55",
  publisher="IEEE",
  address="Ghent, Belgium",
  doi="10.1109/ICUMT61075.2023.10333096",
  isbn="979-8-3503-9328-6",
  url="https://ieeexplore.ieee.org/document/10333096"
}