Alzheimer’s Disease Diagnosis Using Parallel Convolutional Neural Networks

Alzheimer's disease is progressive brain cell damage. This damage can be slowed down if the disease is detected at an early stage. Computer-Aided detection systems bring great help for this problem especially with the development of convolution neural networks. This work proposes an Alzheimer's disease diagnostic approach based on a segmentation technique and convolution neural networks. We segment an MRI brain image by a recursive thresholding process. This segmentation generates a set of output images witch simulate progressive brain degradation. Convolution neural networks capture this degradation and assign a value that indicates Alzheimer's disease stage.