Application Of Machine Learning Methods in Diagnosis of Alzheimer Disease Based on Fractal Feature Extraction and Convolutional Neural Network

Alzheimer's disease impairs one's capacity to make sound strategic and operational decisions in authentic settings. It may be harder to adhere efficiently to ordinary concerns such as a stove that is burning or unexpected driving scenarios. Alzheimer's disease is characterized by memory loss. The inability to recall recent events or discussions is one of the first indicators. Memory problems intensify as the disease develops, and additional symptoms emerge. This research presents a CNN-based architecture with specific processes starting with image acquisition and ending with AD-classification to categorize scanned MRI images in order to predict whether or not they contain Alzheimer's disease utilizing a machine learning application and digital image processing. Six machine learning algorithms for identifying and segmenting MRI scans of patients' brains are presented in this paper. Six machine learning approaches were used to identify and segment ultrasound pictures in order to determine the illness or tumor kind. First, the fractal approach is used to extract the picture characteristics. The photos of the individuals were then classified using KNN, SVM, DT, and NB classification algorithms. The suggested high-potential CNN methodology may be utilized to identify AD using MRI scans, according to the results of this study.