Objective: Alzheimer's disease (AD) is a chronic disease that causes the death of nerve cells and tissue loss in the brain. It usually starts slowly and worsens over time. Individual computer aided systems are needed for early and accurate diagnosis of Alzheimer's. Magnetic resonance imaging (MRI) offers the opportunity to examine the pathological brain changes associated with AD. In recent years, neuroimaging data has been increasingly used to characterize AD with machine learning methods, offering promising tools for personalized diagnosis. Very recently, a number of studies have proposed to aid the diagnosis of AD through convolutional neural networks (CNNs).Methods: CNN is machine learning algorithm which is used in a variety of fields, including image and pattern recognition, speech recognition, natural language processing, and video analysis. In this study Discrete wavelet transform (DWT) was used for feature extraction. DWT has attractive properties and has better image representation than other transforms like Fourier. Medical images are vulnerable to noise, they are preprocessed to remove unwanted data and improve quality. Feature extraction and classification are two essential components for the recognition system that have a significant impact on the efficiency of the system. DWT is an implementation of wavelet transform that uses a separate set of wavelet scales and translations that follow some defined rules. The aim of this study is to detect Alzheimer's disease by using convolutional neural networks and to reduce noise by preprocessing by applying DWT on the entered images.Results: With combining DWT feature extraction and CNN algorithm for detecting Alzheimer's disease, the performance and learning rate are significantly decreased. The accuracy of the model results based on pure CNN with machine learning algorithm is higher than with than the accuracy in CNN learning without DWT future extraction. Accuracy values are 75% and 69 % successively.
Conclusion:Ultimately, this study revealed that the combination of MET-PLGA NPs with current cancer therapies holds promise for the potential of breast cancer treatment.