Deep Learning Model to Evaluate Alzheimer's disease Through Multi-View Clustering

Early diagnosis of Alzheimer's disease (AD) plays a crucial role in the development and effectiveness of interventions, and neuroimaging stands out as an up-and-coming field for the initial identification of Alzheimer's disease. Earlier models utilized various methods to analyze images of Alzheimer's disease, such as deep learning models or unsupervised matrix factorization processes. Neither of these techniques alone can produce satisfactory results while clustering multi-view photos of Alzheimer's disease. This motivates our research to create a deep learning model for obtaining the most important Alzheimer's disease factors from MRI and classifying brain images into different stages. To achieve optimal results in multi-view clustering, the proposed model integrates a deep learning technique (Channel Boost-Convolution Neural Network) with an inverse matrix factorization method, forming an ensemble approach. The experiment analyzes several images to evaluate the implemented technique for the performance of RMSE, which are about 2.32 better than the various compared models. The results show that combining the deep learning model with Inverse matrix factorization for Alzheimer's disease multi-view image clustering works well, the Transformers can further improve multi-view clustering in deep learning.