Comparison of the Function of the Elman Neural Network and the Deep Neural Network for the Diagnosis of Mild Alzheimer's Disease
The main purpose of this study is to provide a method for early diagnosis of Alzheimer's disease. The disease reduces memory function by destroying nerve cells in the nervous system and reducing nerve connections and interactions.
The level of the disease should be diagnosed according to the association of the disease with various features in the brain signal and medical images. First, with proper preprocessing, nonlinear properties such as phase diagram, correlation dimension, entropy and Lyapunov exponent are extracted and Elman neural network is used for classification. Then, the correctness of the function of Elman neural network is compared with channel neural network. The use of deep learning methods, including channel neural network, can have more appropriate and accurate results among other classification methods.
In the case of using two CNN networks and one MLP network, the accuracy of the results was %98 in healthy individuals, %97.7 in mild patients and %97.5 in severely ill patients. In the case of using a CNN network with a combination of features, brain signal and medical images, in the case of stimulation, the accuracy of the results is %95 in healthy individuals, %92.5 in mild patients and %97.5 in severe patients. As a recall, the accuracy of the results is %75 in healthy individuals, %72.5 in mild patients and %87.5 in severe patients. The accuracy of the results in Elman neural network with the combination of brain signal features and medical images is %94.4 and in the case without combination of features, the accuracy of the results is %92.2.
Among the processing methods proposed to classify the three classes of healthy, mild patient and severe patient, the method of combining brain signal characteristics and medical images has increased the accuracy of CNN and Elman classifier results.
- حق عضویت دریافتی صرف حمایت از نشریات عضو و نگهداری، تکمیل و توسعه مگیران میشود.
- پرداخت حق اشتراک و دانلود مقالات اجازه بازنشر آن در سایر رسانههای چاپی و دیجیتال را به کاربر نمیدهد.