Multi class Alzheimer disease detection using deep learning techniques | IEEE Conference Publication | IEEE Xplore

Multi class Alzheimer disease detection using deep learning techniques


Abstract:

Alzheimer's disease (AD) is a neurodegenerative disorder. Alzheimer's Disease is a predominant kind of memory disorder. Alzheimer's Disease is not reversible, and progres...Show More

Abstract:

Alzheimer's disease (AD) is a neurodegenerative disorder. Alzheimer's Disease is a predominant kind of memory disorder. Alzheimer's Disease is not reversible, and progressive brain disease shows a decrease in thinking with no validated disease-modifying treatment, implying no cure. The Disease's side effects are harmless initially, but later they become more extreme over the long run. Thus, a great effort is needed to develop methods for detecting early, especially at stages like very mild, mild, moderate, to slow or prevent disease progression because, on the off chance that the sickness is anticipated before, the Doctor can slow down the cell degeneration. To this end, the construction of a good prediction system for Alzheimer's Disease is the paper's aim, often reducing time to treatment, medical errors, and overall healthcare cost. This paper utilizes AI calculations to anticipate Alzheimer's illness utilizing Brain MRI Scans. In our study, we trained CNN models to detect from brain MRI images. The MRI image dataset has been collected from The OASIS dataset. Our study has three different Deep CNN models: VGG - 16, Inception-V3, and Xception to classify Alzheimer's Disease. We trained and evaluated all the Deep CNN models. After a relatively short amount of training (125 epochs), we achieved a 75% accuracy with the VGG-16 model, 70% accuracy with the Inception-V3 model, and 70% with the Xception model. Thus, we got very high classification accuracy with our research. In this paper, categorizing Alzheimer's disorder is done into multiple classes (no-dementia, very mild, mild, moderate) to enable the patient to undergo the best and most efficient treatment plan to the patient right away.
Date of Conference: 23-25 March 2022
Date Added to IEEE Xplore: 02 May 2022
ISBN Information:
Conference Location: Chiangrai, Thailand

Contact IEEE to Subscribe

References

References is not available for this document.