Application of Fusion Model of 3D-ResNeXt and Bi-LSTM Network in Alzheimer’s Disease Classification

Alzheimer’s disease is a degenerative disease of the nervous system. If the doctor can detect the disease early, he can treat the patient in advance to slow down the deterioration of the health. We propose a network 3D_ResNeXt_Bi-LSTM fused with ResNeXt and Bi-LSTM, which uses MRI brain images to classify and recognize AD (Alzheimer disease) and NC (Normal Contrast) from neuroimaging. We use a 3D convolution kernel to replace the 2D convolution kernel and flatten the feature of the final ResNeXt into one-dimensional data and send it to Bi-LSTM. So that the network can thoroughly learn the spatial information of the 3D brain image data, finally we send the features to the classifier for classification. Experiments on the ADNI dataset show that our network’s highest classification accuracy for AD and NC is 98.97%.