Detection of Alzheimer’s Disease in Resonance Imaging Images Using Machine Learning with the Convolution Neural Network Method

Alzheimer’s disease is one of the most common types on disease in the elderly, which was characterizing by damage to the brain and can affect memory capacity, speech and behavior. One way to detect Alzheimer’s disease is to use magnetic resonance imaging (MRI). However, the proses of detecting using MRI images still requires manual interpretation which is time consuming and difficult for all doctors to do. Therefore, we need a method that can help simplify and speed up the process of detecting Alzheimer’s disease. Machine learning, especially the Convolution Neural Network (CNN) method, is a method that has the ability to manage and analyze large and complex data and take 2D images as input. In this research, the CNN method was using with the efficienctNetB0 architectural model. Then the model that has been train will be implementing into a simple web application. The results of this study managed to achieve a very high level of accuracy, which is equal to 99%. In addition, the model in this research is able to provide consistent and stable results. This research shows that the use of CNN with the EfficientB0 model is very effective in detecting Alzheimer’s disease on MRI images. These high accuracy results demonstrate the important potential of machine learning approaches in the early diagnosis of this disease. It was hoped that this research can contribute to the development of more accurate and efficient diagnostics aids, which can improve the management and care of Alzheimer’s disease in patients in the future.