Diagnosis of Alzheimer's disease with Convolutional neural network from MRI

mazrooei rad, elias; Mazinani, Sayyed Majid; akbari, hadi

Volume 25, Issue 4 (Winter 2024) Advances in Cognitive Sciences 2024, 25(4): 90-98 | Back to browse issues page

Ethics code: IR.IMAMREZA.REC.1402.009

Mendeley

Zotero

RefWorks

mazrooei rad E, Mazinani S M, akbari H. Diagnosis of Alzheimer's disease with Convolutional neural network from MRI. Advances in Cognitive Sciences 2024; 25 (4) :90-98
URL: http://icssjournal.ir/article-1-1620-en.html

Diagnosis of Alzheimer's disease with Convolutional neural network from MRI

Elias Mazrooei rad ^*

¹, Sayyed Majid Mazinani²

, Hadi Akbari¹

1- khavaran higher education institute
2- Imam reza international University

Abstract: (173 Views)

This paper presents a new method for diagnosing Alzheimer's disease based on the characteristics of MRI images. MRI images are taken with a minimum of 3 Tesla and a thickness of 3 mm to determine aging plaques and spiral coils. Methods: the characteristics of MRI images such as Medial temporal lobe atrophy, white matter volume, gray matter volume, cerebrospinal fluid and asymmetry are determined. The subjects were divided into three groups: healthy people, mild and severe patients. The asymmetry and the mean rate of temporal lobe atrophy increase with the progression of Alzheimer's disease because the amount of damage to the temporal lobe in MRI images of Alzheimer's disease has increased. The accuracy of the Elman's recurrent neural results with the features extracted from the MRI images is compared with the accuracy of the convolutional neural network results. The accuracy of the results by combining the features in healthy individuals was 82.5%. It is 86.5% in mild Alzheimer's patients and 94.5% in severe Alzheimer's patients. The highest accuracy of progeny in the group of severe Alzheimer's patients and the most appropriate feature among the features of MRI images is the degree of Medial temporal lobe atrophy. The use of convolutional neural network shows that the accuracy of the results is 98% in the healthy group, 97.7% in the mild group and 97.5% in the severe patient group. These results show the performance of the convolutional neural network in comparison with the Elman has higher accuracy results

Keywords: Alzheimer's disease, MRI images, Medial temporal lobe atrophy, asymmetry, Recursive neural network

Type of Study: Research |
Received: 2023/09/30 | Accepted: 2024/03/11 | Published: 2024/04/22

Send email to the article author

Rights and permissions
	This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Designed & Developed by : Yektaweb